/[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 150 by ph10, Tue Apr 17 08:22:40 2007 UTC revision 1194 by ph10, Wed Oct 31 17:42:29 2012 UTC
# Line 1  Line 1 
1  .TH PCREAPI 3  .TH PCREAPI 3 "31 October 2012" "PCRE 8.32"
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
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
# Line 25  PCRE - Perl-compatible regular expressio Line 27  PCRE - Perl-compatible regular expressio
27  .ti +5n  .ti +5n
28  .B const char **\fIerrptr\fP);  .B const char **\fIerrptr\fP);
29  .PP  .PP
30    .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,
# Line 38  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    .
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 82  PCRE - Perl-compatible regular expressio Line 90  PCRE - Perl-compatible regular expressio
90  .B void pcre_free_substring(const char *\fIstringptr\fP);  .B void pcre_free_substring(const char *\fIstringptr\fP);
91  .PP  .PP
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 int pcre_jit_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
99    .ti +5n
100    .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
101    .ti +5n
102    .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP,
103    .ti +5n
104    .B pcre_jit_stack *\fIjstack\fP);
105    .PP
106    .B pcre_jit_stack *pcre_jit_stack_alloc(int \fIstartsize\fP, int \fImaxsize\fP);
107    .PP
108    .B void pcre_jit_stack_free(pcre_jit_stack *\fIstack\fP);
109    .PP
110    .B void pcre_assign_jit_stack(pcre_extra *\fIextra\fP,
111    .ti +5n
112    .B pcre_jit_callback \fIcallback\fP, void *\fIdata\fP);
113  .PP  .PP
114  .B const unsigned char *pcre_maketables(void);  .B const unsigned char *pcre_maketables(void);
115  .PP  .PP
# Line 89  PCRE - Perl-compatible regular expressio Line 117  PCRE - Perl-compatible regular expressio
117  .ti +5n  .ti +5n
118  .B int \fIwhat\fP, void *\fIwhere\fP);  .B int \fIwhat\fP, void *\fIwhere\fP);
119  .PP  .PP
 .B int pcre_info(const pcre *\fIcode\fP, int *\fIoptptr\fP, int  
 .B *\fIfirstcharptr\fP);  
 .PP  
120  .B int pcre_refcount(pcre *\fIcode\fP, int \fIadjust\fP);  .B int pcre_refcount(pcre *\fIcode\fP, int \fIadjust\fP);
121  .PP  .PP
122  .B int pcre_config(int \fIwhat\fP, void *\fIwhere\fP);  .B int pcre_config(int \fIwhat\fP, void *\fIwhere\fP);
123  .PP  .PP
124  .B char *pcre_version(void);  .B const char *pcre_version(void);
125  .PP  .PP
126    .B int pcre_pattern_to_host_byte_order(pcre *\fIcode\fP,
127    .ti +5n
128    .B pcre_extra *\fIextra\fP, const unsigned char *\fItables\fP);
129    .
130    .
131    .SH "PCRE NATIVE API INDIRECTED FUNCTIONS"
132    .rs
133    .sp
134  .B void *(*pcre_malloc)(size_t);  .B void *(*pcre_malloc)(size_t);
135  .PP  .PP
136  .B void (*pcre_free)(void *);  .B void (*pcre_free)(void *);
# Line 109  PCRE - Perl-compatible regular expressio Line 142  PCRE - Perl-compatible regular expressio
142  .B int (*pcre_callout)(pcre_callout_block *);  .B int (*pcre_callout)(pcre_callout_block *);
143  .  .
144  .  .
145    .SH "PCRE 8-BIT, 16-BIT AND 32-BIT LIBRARIES"
146    .rs
147    .sp
148    From release 8.30, PCRE can be compiled as a library for handling 16-bit
149    character strings as well as, or instead of, the original library that handles
150    8-bit character strings. From release 8.32, PCRE can also be compiled as a
151    library for handling 32-bit character strings. To avoid too much complication,
152    this document describes the 8-bit versions of the functions, with only
153    occasional references to the 16-bit and 32-bit libraries.
154    .P
155    The 16-bit and 32-bit functions operate in the same way as their 8-bit
156    counterparts; they just use different data types for their arguments and
157    results, and their names start with \fBpcre16_\fP or \fBpcre32_\fP instead of
158    \fBpcre_\fP. For every option that has UTF8 in its name (for example,
159    PCRE_UTF8), there are corresponding 16-bit and 32-bit names with UTF8 replaced
160    by UTF16 or UTF32, respectively. This facility is in fact just cosmetic; the
161    16-bit and 32-bit option names define the same bit values.
162    .P
163    References to bytes and UTF-8 in this document should be read as references to
164    16-bit data quantities and UTF-16 when using the 16-bit library, or 32-bit data
165    quantities and UTF-32 when using the 32-bit library, unless specified
166    otherwise. More details of the specific differences for the 16-bit and 32-bit
167    libraries are given in the
168    .\" HREF
169    \fBpcre16\fP
170    .\"
171    and
172    .\" HREF
173    \fBpcre32\fP
174    .\"
175    pages.
176    .
177    .
178  .SH "PCRE API OVERVIEW"  .SH "PCRE API OVERVIEW"
179  .rs  .rs
180  .sp  .sp
181  PCRE has its own native API, which is described in this document. There are  PCRE has its own native API, which is described in this document. There are
182  also some wrapper functions that correspond to the POSIX regular expression  also some wrapper functions (for the 8-bit library only) that correspond to the
183  API. These are described in the  POSIX regular expression API, but they do not give access to all the
184    functionality. They are described in the
185  .\" HREF  .\" HREF
186  \fBpcreposix\fP  \fBpcreposix\fP
187  .\"  .\"
188  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++
189  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
190    documented in the
191  .\" HREF  .\" HREF
192  \fBpcrecpp\fP  \fBpcrecpp\fP
193  .\"  .\"
194  page.  page.
195  .P  .P
196  The native API C function prototypes are defined in the header file  The native API C function prototypes are defined in the header file
197  \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
198  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
199  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
200  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
201  Applications can use these to include support for different releases of PCRE.  for the library. Applications can use these to include support for different
202    releases of PCRE.
203    .P
204    In a Windows environment, if you want to statically link an application program
205    against a non-dll \fBpcre.a\fP file, you must define PCRE_STATIC before
206    including \fBpcre.h\fP or \fBpcrecpp.h\fP, because otherwise the
207    \fBpcre_malloc()\fP and \fBpcre_free()\fP exported functions will be declared
208    \fB__declspec(dllimport)\fP, with unwanted results.
209  .P  .P
210  The functions \fBpcre_compile()\fP, \fBpcre_compile2()\fP, \fBpcre_study()\fP,  The functions \fBpcre_compile()\fP, \fBpcre_compile2()\fP, \fBpcre_study()\fP,
211  and \fBpcre_exec()\fP are used for compiling and matching regular expressions  and \fBpcre_exec()\fP are used for compiling and matching regular expressions
212  in a Perl-compatible manner. A sample program that demonstrates the simplest  in a Perl-compatible manner. A sample program that demonstrates the simplest
213  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
214  distribution. The  source distribution. A listing of this program is given in the
215    .\" HREF
216    \fBpcredemo\fP
217    .\"
218    documentation, and the
219  .\" HREF  .\" HREF
220  \fBpcresample\fP  \fBpcresample\fP
221  .\"  .\"
222  documentation describes how to run it.  documentation describes how to compile and run it.
223    .P
224    Just-in-time compiler support is an optional feature of PCRE that can be built
225    in appropriate hardware environments. It greatly speeds up the matching
226    performance of many patterns. Simple programs can easily request that it be
227    used if available, by setting an option that is ignored when it is not
228    relevant. More complicated programs might need to make use of the functions
229    \fBpcre_jit_stack_alloc()\fP, \fBpcre_jit_stack_free()\fP, and
230    \fBpcre_assign_jit_stack()\fP in order to control the JIT code's memory usage.
231    .P
232    From release 8.32 there is also a direct interface for JIT execution, which
233    gives improved performance. The JIT-specific functions are discussed in the
234    .\" HREF
235    \fBpcrejit\fP
236    .\"
237    documentation.
238  .P  .P
239  A second matching function, \fBpcre_dfa_exec()\fP, which is not  A second matching function, \fBpcre_dfa_exec()\fP, which is not
240  Perl-compatible, is also provided. This uses a different algorithm for the  Perl-compatible, is also provided. This uses a different algorithm for the
241  matching. The alternative algorithm finds all possible matches (at a given  matching. The alternative algorithm finds all possible matches (at a given
242  point in the subject), and scans the subject just once. However, this algorithm  point in the subject), and scans the subject just once (unless there are
243  does not return captured substrings. A description of the two matching  lookbehind assertions). However, this algorithm does not return captured
244  algorithms and their advantages and disadvantages is given in the  substrings. A description of the two matching algorithms and their advantages
245    and disadvantages is given in the
246  .\" HREF  .\" HREF
247  \fBpcrematching\fP  \fBpcrematching\fP
248  .\"  .\"
# Line 175  specialist use. Most commonly, no specia Line 270  specialist use. Most commonly, no specia
270  internal tables that are generated when PCRE is built are used.  internal tables that are generated when PCRE is built are used.
271  .P  .P
272  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
273  compiled pattern; \fBpcre_info()\fP is an obsolete version that returns only  compiled pattern. The function \fBpcre_version()\fP returns a pointer to a
274  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.  
275  .P  .P
276  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
277  containing a compiled pattern. This is provided for the benefit of  containing a compiled pattern. This is provided for the benefit of
# Line 218  points during a matching operation. Deta Line 311  points during a matching operation. Deta
311  documentation.  documentation.
312  .  .
313  .  .
314    .\" HTML <a name="newlines"></a>
315  .SH NEWLINES  .SH NEWLINES
316  .rs  .rs
317  .sp  .sp
# Line 225  PCRE supports five different conventions Line 319  PCRE supports five different conventions
319  strings: a single CR (carriage return) character, a single LF (linefeed)  strings: a single CR (carriage return) character, a single LF (linefeed)
320  character, the two-character sequence CRLF, any of the three preceding, or any  character, the two-character sequence CRLF, any of the three preceding, or any
321  Unicode newline sequence. The Unicode newline sequences are the three just  Unicode newline sequence. The Unicode newline sequences are the three just
322  mentioned, plus the single characters VT (vertical tab, U+000B), FF (formfeed,  mentioned, plus the single characters VT (vertical tab, U+000B), FF (form feed,
323  U+000C), NEL (next line, U+0085), LS (line separator, U+2028), and PS  U+000C), NEL (next line, U+0085), LS (line separator, U+2028), and PS
324  (paragraph separator, U+2029).  (paragraph separator, U+2029).
325  .P  .P
# Line 235  The default default is LF, which is the Line 329  The default default is LF, which is the
329  default can be overridden, either when a pattern is compiled, or when it is  default can be overridden, either when a pattern is compiled, or when it is
330  matched.  matched.
331  .P  .P
332    At compile time, the newline convention can be specified by the \fIoptions\fP
333    argument of \fBpcre_compile()\fP, or it can be specified by special text at the
334    start of the pattern itself; this overrides any other settings. See the
335    .\" HREF
336    \fBpcrepattern\fP
337    .\"
338    page for details of the special character sequences.
339    .P
340  In the PCRE documentation the word "newline" is used to mean "the character or  In the PCRE documentation the word "newline" is used to mean "the character or
341  pair of characters that indicate a line break". The choice of newline  pair of characters that indicate a line break". The choice of newline
342  convention affects the handling of the dot, circumflex, and dollar  convention affects the handling of the dot, circumflex, and dollar
343  metacharacters, the handling of #-comments in /x mode, and, when CRLF is a  metacharacters, the handling of #-comments in /x mode, and, when CRLF is a
344  recognized line ending sequence, the match position advancement for a  recognized line ending sequence, the match position advancement for a
345  non-anchored pattern. The choice of newline convention does not affect the  non-anchored pattern. There is more detail about this in the
346  interpretation of the \en or \er escape sequences.  .\" HTML <a href="#execoptions">
347    .\" </a>
348    section on \fBpcre_exec()\fP options
349    .\"
350    below.
351    .P
352    The choice of newline convention does not affect the interpretation of
353    the \en or \er escape sequences, nor does it affect what \eR matches, which is
354    controlled in a similar way, but by separate options.
355  .  .
356  .  .
357  .SH MULTITHREADING  .SH MULTITHREADING
# Line 254  callout function pointed to by \fBpcre_c Line 364  callout function pointed to by \fBpcre_c
364  .P  .P
365  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
366  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.
367    .P
368    If the just-in-time optimization feature is being used, it needs separate
369    memory stack areas for each thread. See the
370    .\" HREF
371    \fBpcrejit\fP
372    .\"
373    documentation for more details.
374  .  .
375  .  .
376  .SH "SAVING PRECOMPILED PATTERNS FOR LATER USE"  .SH "SAVING PRECOMPILED PATTERNS FOR LATER USE"
# Line 265  which it was compiled. Details are given Line 382  which it was compiled. Details are given
382  .\" HREF  .\" HREF
383  \fBpcreprecompile\fP  \fBpcreprecompile\fP
384  .\"  .\"
385  documentation.  documentation, which includes a description of the
386    \fBpcre_pattern_to_host_byte_order()\fP function. However, compiling a regular
387    expression with one version of PCRE for use with a different version is not
388    guaranteed to work and may cause crashes.
389  .  .
390  .  .
391  .SH "CHECKING BUILD-TIME OPTIONS"  .SH "CHECKING BUILD-TIME OPTIONS"
# Line 282  documentation has more details about the Line 402  documentation has more details about the
402  .P  .P
403  The first argument for \fBpcre_config()\fP is an integer, specifying which  The first argument for \fBpcre_config()\fP is an integer, specifying which
404  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
405  which the information is placed. The following information is available:  which the information is placed. The returned value is zero on success, or the
406    negative error code PCRE_ERROR_BADOPTION if the value in the first argument is
407    not recognized. The following information is available:
408  .sp  .sp
409    PCRE_CONFIG_UTF8    PCRE_CONFIG_UTF8
410  .sp  .sp
411  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;
412  otherwise it is set to zero.  otherwise it is set to zero. This value should normally be given to the 8-bit
413    version of this function, \fBpcre_config()\fP. If it is given to the 16-bit
414    or 32-bit version of this function, the result is PCRE_ERROR_BADOPTION.
415    .sp
416      PCRE_CONFIG_UTF16
417    .sp
418    The output is an integer that is set to one if UTF-16 support is available;
419    otherwise it is set to zero. This value should normally be given to the 16-bit
420    version of this function, \fBpcre16_config()\fP. If it is given to the 8-bit
421    or 32-bit version of this function, the result is PCRE_ERROR_BADOPTION.
422    .sp
423      PCRE_CONFIG_UTF32
424    .sp
425    The output is an integer that is set to one if UTF-32 support is available;
426    otherwise it is set to zero. This value should normally be given to the 32-bit
427    version of this function, \fBpcre32_config()\fP. If it is given to the 8-bit
428    or 16-bit version of this function, the result is PCRE_ERROR_BADOPTION.
429  .sp  .sp
430    PCRE_CONFIG_UNICODE_PROPERTIES    PCRE_CONFIG_UNICODE_PROPERTIES
431  .sp  .sp
432  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
433  properties is available; otherwise it is set to zero.  properties is available; otherwise it is set to zero.
434  .sp  .sp
435      PCRE_CONFIG_JIT
436    .sp
437    The output is an integer that is set to one if support for just-in-time
438    compiling is available; otherwise it is set to zero.
439    .sp
440      PCRE_CONFIG_JITTARGET
441    .sp
442    The output is a pointer to a zero-terminated "const char *" string. If JIT
443    support is available, the string contains the name of the architecture for
444    which the JIT compiler is configured, for example "x86 32bit (little endian +
445    unaligned)". If JIT support is not available, the result is NULL.
446    .sp
447    PCRE_CONFIG_NEWLINE    PCRE_CONFIG_NEWLINE
448  .sp  .sp
449  The output is an integer whose value specifies the default character sequence  The output is an integer whose value specifies the default character sequence
450  that is recognized as meaning "newline". The four values that are supported  that is recognized as meaning "newline". The values that are supported in
451  are: 10 for LF, 13 for CR, 3338 for CRLF, -2 for ANYCRLF, and -1 for ANY. The  ASCII/Unicode environments are: 10 for LF, 13 for CR, 3338 for CRLF, -2 for
452  default should normally be the standard sequence for your operating system.  ANYCRLF, and -1 for ANY. In EBCDIC environments, CR, ANYCRLF, and ANY yield the
453    same values. However, the value for LF is normally 21, though some EBCDIC
454    environments use 37. The corresponding values for CRLF are 3349 and 3365. The
455    default should normally correspond to the standard sequence for your operating
456    system.
457    .sp
458      PCRE_CONFIG_BSR
459    .sp
460    The output is an integer whose value indicates what character sequences the \eR
461    escape sequence matches by default. A value of 0 means that \eR matches any
462    Unicode line ending sequence; a value of 1 means that \eR matches only CR, LF,
463    or CRLF. The default can be overridden when a pattern is compiled or matched.
464  .sp  .sp
465    PCRE_CONFIG_LINK_SIZE    PCRE_CONFIG_LINK_SIZE
466  .sp  .sp
467  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
468  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
469  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
470  matching. The default value of 2 is sufficient for all but the most massive  a number of bytes. For the 32-bit library, the value is either 2 or 4 and is
471  patterns, since it allows the compiled pattern to be up to 64K in size.  still a number of bytes. The default value of 2 is sufficient for all but the
472    most massive patterns, since it allows the compiled pattern to be up to 64K in
473    size. Larger values allow larger regular expressions to be compiled, at the
474    expense of slower matching.
475  .sp  .sp
476    PCRE_CONFIG_POSIX_MALLOC_THRESHOLD    PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
477  .sp  .sp
# Line 321  documentation. Line 485  documentation.
485  .sp  .sp
486    PCRE_CONFIG_MATCH_LIMIT    PCRE_CONFIG_MATCH_LIMIT
487  .sp  .sp
488  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
489  internal matching function calls in a \fBpcre_exec()\fP execution. Further  internal matching function calls in a \fBpcre_exec()\fP execution. Further
490  details are given with \fBpcre_exec()\fP below.  details are given with \fBpcre_exec()\fP below.
491  .sp  .sp
492    PCRE_CONFIG_MATCH_LIMIT_RECURSION    PCRE_CONFIG_MATCH_LIMIT_RECURSION
493  .sp  .sp
494  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
495  recursion when calling the internal matching function in a \fBpcre_exec()\fP  recursion when calling the internal matching function in a \fBpcre_exec()\fP
496  execution. Further details are given with \fBpcre_exec()\fP below.  execution. Further details are given with \fBpcre_exec()\fP below.
497  .sp  .sp
# Line 362  avoiding the use of the stack. Line 526  avoiding the use of the stack.
526  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
527  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
528  the two interfaces is that \fBpcre_compile2()\fP has an additional argument,  the two interfaces is that \fBpcre_compile2()\fP has an additional argument,
529  \fIerrorcodeptr\fP, via which a numerical error code can be returned.  \fIerrorcodeptr\fP, via which a numerical error code can be returned. To avoid
530    too much repetition, we refer just to \fBpcre_compile()\fP below, but the
531    information applies equally to \fBpcre_compile2()\fP.
532  .P  .P
533  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
534  \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
# Line 378  argument, which is an address (see below Line 544  argument, which is an address (see below
544  .P  .P
545  The \fIoptions\fP argument contains various bit settings that affect the  The \fIoptions\fP argument contains various bit settings that affect the
546  compilation. It should be zero if no options are required. The available  compilation. It should be zero if no options are required. The available
547  options are described below. Some of them, in particular, those that are  options are described below. Some of them (in particular, those that are
548  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
549  the detailed description in the  within the pattern (see the detailed description in the
550  .\" HREF  .\" HREF
551  \fBpcrepattern\fP  \fBpcrepattern\fP
552  .\"  .\"
553  documentation). For these options, the contents of the \fIoptions\fP argument  documentation). For those options that can be different in different parts of
554  specifies their initial settings at the start of compilation and execution. The  the pattern, the contents of the \fIoptions\fP argument specifies their
555  PCRE_ANCHORED and PCRE_NEWLINE_\fIxxx\fP options can be set at the time of  settings at the start of compilation and execution. The PCRE_ANCHORED,
556  matching as well as at compile time.  PCRE_BSR_\fIxxx\fP, PCRE_NEWLINE_\fIxxx\fP, PCRE_NO_UTF8_CHECK, and
557    PCRE_NO_START_OPTIMIZE options can be set at the time of matching as well as at
558    compile time.
559  .P  .P
560  If \fIerrptr\fP is NULL, \fBpcre_compile()\fP returns NULL immediately.  If \fIerrptr\fP is NULL, \fBpcre_compile()\fP returns NULL immediately.
561  Otherwise, if compilation of a pattern fails, \fBpcre_compile()\fP returns  Otherwise, if compilation of a pattern fails, \fBpcre_compile()\fP returns
562  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
563  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
564  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
565  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
566  \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
567    immediate error is given). However, for an invalid UTF-8 string, the offset is
568    that of the first byte of the failing character.
569    .P
570    Some errors are not detected until the whole pattern has been scanned; in these
571    cases, the offset passed back is the length of the pattern. Note that the
572    offset is in bytes, not characters, even in UTF-8 mode. It may sometimes point
573    into the middle of a UTF-8 character.
574  .P  .P
575  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
576  \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 442  facility, see the Line 617  facility, see the
617  .\"  .\"
618  documentation.  documentation.
619  .sp  .sp
620      PCRE_BSR_ANYCRLF
621      PCRE_BSR_UNICODE
622    .sp
623    These options (which are mutually exclusive) control what the \eR escape
624    sequence matches. The choice is either to match only CR, LF, or CRLF, or to
625    match any Unicode newline sequence. The default is specified when PCRE is
626    built. It can be overridden from within the pattern, or by setting an option
627    when a compiled pattern is matched.
628    .sp
629    PCRE_CASELESS    PCRE_CASELESS
630  .sp  .sp
631  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 465  pattern. Line 649  pattern.
649  .sp  .sp
650    PCRE_DOTALL    PCRE_DOTALL
651  .sp  .sp
652  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
653  including those that indicate newline. Without it, a dot does not match when  any value, including one that indicates a newline. However, it only ever
654  the current position is at a newline. This option is equivalent to Perl's /s  matches one character, even if newlines are coded as CRLF. Without this option,
655  option, and it can be changed within a pattern by a (?s) option setting. A  a dot does not match when the current position is at a newline. This option is
656  negative class such as [^a] always matches newline characters, independent of  equivalent to Perl's /s option, and it can be changed within a pattern by a
657  the setting of this option.  (?s) option setting. A negative class such as [^a] always matches newline
658    characters, independent of the setting of this option.
659  .sp  .sp
660    PCRE_DUPNAMES    PCRE_DUPNAMES
661  .sp  .sp
# Line 485  documentation. Line 670  documentation.
670  .sp  .sp
671    PCRE_EXTENDED    PCRE_EXTENDED
672  .sp  .sp
673  If this bit is set, whitespace data characters in the pattern are totally  If this bit is set, white space data characters in the pattern are totally
674  ignored except when escaped or inside a character class. Whitespace does not  ignored except when escaped or inside a character class. White space does not
675  include the VT character (code 11). In addition, characters between an  include the VT character (code 11). In addition, characters between an
676  unescaped # outside a character class and the next newline, inclusive, are also  unescaped # outside a character class and the next newline, inclusive, are also
677  ignored. This is equivalent to Perl's /x option, and it can be changed within a  ignored. This is equivalent to Perl's /x option, and it can be changed within a
678  pattern by a (?x) option setting.  pattern by a (?x) option setting.
679  .P  .P
680    Which characters are interpreted as newlines is controlled by the options
681    passed to \fBpcre_compile()\fP or by a special sequence at the start of the
682    pattern, as described in the section entitled
683    .\" HTML <a href="pcrepattern.html#newlines">
684    .\" </a>
685    "Newline conventions"
686    .\"
687    in the \fBpcrepattern\fP documentation. Note that the end of this type of
688    comment is a literal newline sequence in the pattern; escape sequences that
689    happen to represent a newline do not count.
690    .P
691  This option makes it possible to include comments inside complicated patterns.  This option makes it possible to include comments inside complicated patterns.
692  Note, however, that this applies only to data characters. Whitespace characters  Note, however, that this applies only to data characters. White space characters
693  may never appear within special character sequences in a pattern, for example  may never appear within special character sequences in a pattern, for example
694  within the sequence (?( which introduces a conditional subpattern.  within the sequence (?( that introduces a conditional subpattern.
695  .sp  .sp
696    PCRE_EXTRA    PCRE_EXTRA
697  .sp  .sp
# Line 505  set, any backslash in a pattern that is Line 701  set, any backslash in a pattern that is
701  special meaning causes an error, thus reserving these combinations for future  special meaning causes an error, thus reserving these combinations for future
702  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
703  special meaning is treated as a literal. (Perl can, however, be persuaded to  special meaning is treated as a literal. (Perl can, however, be persuaded to
704  give a warning for this.) There are at present no other features controlled by  give an error for this, by running it with the -w option.) There are at present
705  this option. It can also be set by a (?X) option setting within a pattern.  no other features controlled by this option. It can also be set by a (?X)
706    option setting within a pattern.
707  .sp  .sp
708    PCRE_FIRSTLINE    PCRE_FIRSTLINE
709  .sp  .sp
# Line 514  If this option is set, an unanchored pat Line 711  If this option is set, an unanchored pat
711  the first newline in the subject string, though the matched text may continue  the first newline in the subject string, though the matched text may continue
712  over the newline.  over the newline.
713  .sp  .sp
714      PCRE_JAVASCRIPT_COMPAT
715    .sp
716    If this option is set, PCRE's behaviour is changed in some ways so that it is
717    compatible with JavaScript rather than Perl. The changes are as follows:
718    .P
719    (1) A lone closing square bracket in a pattern causes a compile-time error,
720    because this is illegal in JavaScript (by default it is treated as a data
721    character). Thus, the pattern AB]CD becomes illegal when this option is set.
722    .P
723    (2) At run time, a back reference to an unset subpattern group matches an empty
724    string (by default this causes the current matching alternative to fail). A
725    pattern such as (\e1)(a) succeeds when this option is set (assuming it can find
726    an "a" in the subject), whereas it fails by default, for Perl compatibility.
727    .P
728    (3) \eU matches an upper case "U" character; by default \eU causes a compile
729    time error (Perl uses \eU to upper case subsequent characters).
730    .P
731    (4) \eu matches a lower case "u" character unless it is followed by four
732    hexadecimal digits, in which case the hexadecimal number defines the code point
733    to match. By default, \eu causes a compile time error (Perl uses it to upper
734    case the following character).
735    .P
736    (5) \ex matches a lower case "x" character unless it is followed by two
737    hexadecimal digits, in which case the hexadecimal number defines the code point
738    to match. By default, as in Perl, a hexadecimal number is always expected after
739    \ex, but it may have zero, one, or two digits (so, for example, \exz matches a
740    binary zero character followed by z).
741    .sp
742    PCRE_MULTILINE    PCRE_MULTILINE
743  .sp  .sp
744  By default, PCRE treats the subject string as consisting of a single line of  By default, PCRE treats the subject string as consisting of a single line of
# Line 542  indicated by a single character (CR or L Line 767  indicated by a single character (CR or L
767  PCRE_NEWLINE_CRLF specifies that a newline is indicated by the two-character  PCRE_NEWLINE_CRLF specifies that a newline is indicated by the two-character
768  CRLF sequence. Setting PCRE_NEWLINE_ANYCRLF specifies that any of the three  CRLF sequence. Setting PCRE_NEWLINE_ANYCRLF specifies that any of the three
769  preceding sequences should be recognized. Setting PCRE_NEWLINE_ANY specifies  preceding sequences should be recognized. Setting PCRE_NEWLINE_ANY specifies
770  that any Unicode newline sequence should be recognized. The Unicode newline  that any Unicode newline sequence should be recognized.
771  sequences are the three just mentioned, plus the single characters VT (vertical  .P
772  tab, U+000B), FF (formfeed, U+000C), NEL (next line, U+0085), LS (line  In an ASCII/Unicode environment, the Unicode newline sequences are the three
773  separator, U+2028), and PS (paragraph separator, U+2029). The last two are  just mentioned, plus the single characters VT (vertical tab, U+000B), FF (form
774    feed, U+000C), NEL (next line, U+0085), LS (line separator, U+2028), and PS
775    (paragraph separator, U+2029). For the 8-bit library, the last two are
776  recognized only in UTF-8 mode.  recognized only in UTF-8 mode.
777  .P  .P
778    When PCRE is compiled to run in an EBCDIC (mainframe) environment, the code for
779    CR is 0x0d, the same as ASCII. However, the character code for LF is normally
780    0x15, though in some EBCDIC environments 0x25 is used. Whichever of these is
781    not LF is made to correspond to Unicode's NEL character. EBCDIC codes are all
782    less than 256. For more details, see the
783    .\" HREF
784    \fBpcrebuild\fP
785    .\"
786    documentation.
787    .P
788  The newline setting in the options word uses three bits that are treated  The newline setting in the options word uses three bits that are treated
789  as a number, giving eight possibilities. Currently only six are used (default  as a number, giving eight possibilities. Currently only six are used (default
790  plus the five values above). This means that if you set more than one newline  plus the five values above). This means that if you set more than one newline
# Line 555  option, the combination may or may not b Line 792  option, the combination may or may not b
792  PCRE_NEWLINE_CR with PCRE_NEWLINE_LF is equivalent to PCRE_NEWLINE_CRLF, but  PCRE_NEWLINE_CR with PCRE_NEWLINE_LF is equivalent to PCRE_NEWLINE_CRLF, but
793  other combinations may yield unused numbers and cause an error.  other combinations may yield unused numbers and cause an error.
794  .P  .P
795  The only time that a line break is specially recognized when compiling a  The only time that a line break in a pattern is specially recognized when
796  pattern is if PCRE_EXTENDED is set, and an unescaped # outside a character  compiling is when PCRE_EXTENDED is set. CR and LF are white space characters,
797  class is encountered. This indicates a comment that lasts until after the next  and so are ignored in this mode. Also, an unescaped # outside a character class
798  line break sequence. In other circumstances, line break sequences are treated  indicates a comment that lasts until after the next line break sequence. In
799  as literal data, except that in PCRE_EXTENDED mode, both CR and LF are treated  other circumstances, line break sequences in patterns are treated as literal
800  as whitespace characters and are therefore ignored.  data.
801  .P  .P
802  The newline option that is set at compile time becomes the default that is used  The newline option that is set at compile time becomes the default that is used
803  for \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP, but it can be overridden.  for \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP, but it can be overridden.
# Line 573  were followed by ?: but named parenthese Line 810  were followed by ?: but named parenthese
810  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
811  in Perl.  in Perl.
812  .sp  .sp
813      NO_START_OPTIMIZE
814    .sp
815    This is an option that acts at matching time; that is, it is really an option
816    for \fBpcre_exec()\fP or \fBpcre_dfa_exec()\fP. If it is set at compile time,
817    it is remembered with the compiled pattern and assumed at matching time. For
818    details see the discussion of PCRE_NO_START_OPTIMIZE
819    .\" HTML <a href="#execoptions">
820    .\" </a>
821    below.
822    .\"
823    .sp
824      PCRE_UCP
825    .sp
826    This option changes the way PCRE processes \eB, \eb, \eD, \ed, \eS, \es, \eW,
827    \ew, and some of the POSIX character classes. By default, only ASCII characters
828    are recognized, but if PCRE_UCP is set, Unicode properties are used instead to
829    classify characters. More details are given in the section on
830    .\" HTML <a href="pcre.html#genericchartypes">
831    .\" </a>
832    generic character types
833    .\"
834    in the
835    .\" HREF
836    \fBpcrepattern\fP
837    .\"
838    page. If you set PCRE_UCP, matching one of the items it affects takes much
839    longer. The option is available only if PCRE has been compiled with Unicode
840    property support.
841    .sp
842    PCRE_UNGREEDY    PCRE_UNGREEDY
843  .sp  .sp
844  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 582  with Perl. It can also be set by a (?U) Line 848  with Perl. It can also be set by a (?U)
848    PCRE_UTF8    PCRE_UTF8
849  .sp  .sp
850  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
851  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
852  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
853  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
854  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  
855  .\" HREF  .\" HREF
856  \fBpcre\fP  \fBpcreunicode\fP
857  .\"  .\"
858  page.  page.
859  .sp  .sp
860    PCRE_NO_UTF8_CHECK    PCRE_NO_UTF8_CHECK
861  .sp  .sp
862  When PCRE_UTF8 is set, the validity of the pattern as a UTF-8 string is  When PCRE_UTF8 is set, the validity of the pattern as a UTF-8 string is
863  automatically checked. If an invalid UTF-8 sequence of bytes is found,  automatically checked. There is a discussion about the
864  \fBpcre_compile()\fP returns an error. If you already know that your pattern is  .\" HTML <a href="pcreunicode.html#utf8strings">
865  valid, and you want to skip this check for performance reasons, you can set the  .\" </a>
866  PCRE_NO_UTF8_CHECK option. When it is set, the effect of passing an invalid  validity of UTF-8 strings
867  UTF-8 string as a pattern is undefined. It may cause your program to crash.  .\"
868  Note that this option can also be passed to \fBpcre_exec()\fP and  in the
869  \fBpcre_dfa_exec()\fP, to suppress the UTF-8 validity checking of subject  .\" HREF
870  strings.  \fBpcreunicode\fP
871    .\"
872    page. If an invalid UTF-8 sequence is found, \fBpcre_compile()\fP returns an
873    error. If you already know that your pattern is valid, and you want to skip
874    this check for performance reasons, you can set the PCRE_NO_UTF8_CHECK option.
875    When it is set, the effect of passing an invalid UTF-8 string as a pattern is
876    undefined. It may cause your program to crash. Note that this option can also
877    be passed to \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP, to suppress the
878    validity checking of subject strings only. If the same string is being matched
879    many times, the option can be safely set for the second and subsequent
880    matchings to improve performance.
881  .  .
882  .  .
883  .SH "COMPILATION ERROR CODES"  .SH "COMPILATION ERROR CODES"
# Line 614  strings. Line 885  strings.
885  .sp  .sp
886  The following table lists the error codes than may be returned by  The following table lists the error codes than may be returned by
887  \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
888  both compiling functions. As PCRE has developed, some error codes have fallen  both compiling functions. Note that error messages are always 8-bit ASCII
889  out of use. To avoid confusion, they have not been re-used.  strings, even in 16-bit or 32-bit mode. As PCRE has developed, some error codes
890    have fallen out of use. To avoid confusion, they have not been re-used.
891  .sp  .sp
892     0  no error     0  no error
893     1  \e at end of pattern     1  \e at end of pattern
# Line 629  out of use. To avoid confusion, they hav Line 901  out of use. To avoid confusion, they hav
901     9  nothing to repeat     9  nothing to repeat
902    10  [this code is not in use]    10  [this code is not in use]
903    11  internal error: unexpected repeat    11  internal error: unexpected repeat
904    12  unrecognized character after (?    12  unrecognized character after (? or (?-
905    13  POSIX named classes are supported only within a class    13  POSIX named classes are supported only within a class
906    14  missing )    14  missing )
907    15  reference to non-existent subpattern    15  reference to non-existent subpattern
# Line 637  out of use. To avoid confusion, they hav Line 909  out of use. To avoid confusion, they hav
909    17  unknown option bit(s) set    17  unknown option bit(s) set
910    18  missing ) after comment    18  missing ) after comment
911    19  [this code is not in use]    19  [this code is not in use]
912    20  regular expression too large    20  regular expression is too large
913    21  failed to get memory    21  failed to get memory
914    22  unmatched parentheses    22  unmatched parentheses
915    23  internal error: code overflow    23  internal error: code overflow
# Line 646  out of use. To avoid confusion, they hav Line 918  out of use. To avoid confusion, they hav
918    26  malformed number or name after (?(    26  malformed number or name after (?(
919    27  conditional group contains more than two branches    27  conditional group contains more than two branches
920    28  assertion expected after (?(    28  assertion expected after (?(
921    29  (?R or (?digits must be followed by )    29  (?R or (?[+-]digits must be followed by )
922    30  unknown POSIX class name    30  unknown POSIX class name
923    31  POSIX collating elements are not supported    31  POSIX collating elements are not supported
924    32  this version of PCRE is not compiled with PCRE_UTF8 support    32  this version of PCRE is compiled without UTF support
925    33  [this code is not in use]    33  [this code is not in use]
926    34  character value in \ex{...} sequence is too large    34  character value in \ex{...} sequence is too large
927    35  invalid condition (?(0)    35  invalid condition (?(0)
928    36  \eC not allowed in lookbehind assertion    36  \eC not allowed in lookbehind assertion
929    37  PCRE does not support \eL, \el, \eN, \eU, or \eu    37  PCRE does not support \eL, \el, \eN{name}, \eU, or \eu
930    38  number after (?C is > 255    38  number after (?C is > 255
931    39  closing ) for (?C expected    39  closing ) for (?C expected
932    40  recursive call could loop indefinitely    40  recursive call could loop indefinitely
933    41  unrecognized character after (?P    41  unrecognized character after (?P
934    42  syntax error in subpattern name (missing terminator)    42  syntax error in subpattern name (missing terminator)
935    43  two named subpatterns have the same name    43  two named subpatterns have the same name
936    44  invalid UTF-8 string    44  invalid UTF-8 string (specifically UTF-8)
937    45  support for \eP, \ep, and \eX has not been compiled    45  support for \eP, \ep, and \eX has not been compiled
938    46  malformed \eP or \ep sequence    46  malformed \eP or \ep sequence
939    47  unknown property name after \eP or \ep    47  unknown property name after \eP or \ep
940    48  subpattern name is too long (maximum 32 characters)    48  subpattern name is too long (maximum 32 characters)
941    49  too many named subpatterns (maximum 10,000)    49  too many named subpatterns (maximum 10000)
942    50  repeated subpattern is too long    50  [this code is not in use]
943    51  octal value is greater than \e377 (not in UTF-8 mode)    51  octal value is greater than \e377 in 8-bit non-UTF-8 mode
944    52  internal error: overran compiling workspace    52  internal error: overran compiling workspace
945    53  internal error: previously-checked referenced subpattern not found    53  internal error: previously-checked referenced subpattern
946            not found
947    54  DEFINE group contains more than one branch    54  DEFINE group contains more than one branch
948    55  repeating a DEFINE group is not allowed    55  repeating a DEFINE group is not allowed
949    56  inconsistent NEWLINE options"    56  inconsistent NEWLINE options
950      57  \eg is not followed by a braced, angle-bracketed, or quoted
951            name/number or by a plain number
952      58  a numbered reference must not be zero
953      59  an argument is not allowed for (*ACCEPT), (*FAIL), or (*COMMIT)
954      60  (*VERB) not recognized
955      61  number is too big
956      62  subpattern name expected
957      63  digit expected after (?+
958      64  ] is an invalid data character in JavaScript compatibility mode
959      65  different names for subpatterns of the same number are
960            not allowed
961      66  (*MARK) must have an argument
962      67  this version of PCRE is not compiled with Unicode property
963            support
964      68  \ec must be followed by an ASCII character
965      69  \ek is not followed by a braced, angle-bracketed, or quoted name
966      70  internal error: unknown opcode in find_fixedlength()
967      71  \eN is not supported in a class
968      72  too many forward references
969      73  disallowed Unicode code point (>= 0xd800 && <= 0xdfff)
970      74  invalid UTF-16 string (specifically UTF-16)
971      75  name is too long in (*MARK), (*PRUNE), (*SKIP), or (*THEN)
972      76  character value in \eu.... sequence is too large
973      77  invalid UTF-32 string (specifically UTF-32)
974    .sp
975    The numbers 32 and 10000 in errors 48 and 49 are defaults; different values may
976    be used if the limits were changed when PCRE was built.
977  .  .
978  .  .
979    .\" HTML <a name="studyingapattern"></a>
980  .SH "STUDYING A PATTERN"  .SH "STUDYING A PATTERN"
981  .rs  .rs
982  .sp  .sp
# Line 692  help speed up matching, \fBpcre_study()\ Line 993  help speed up matching, \fBpcre_study()\
993  results of the study.  results of the study.
994  .P  .P
995  The returned value from \fBpcre_study()\fP can be passed directly to  The returned value from \fBpcre_study()\fP can be passed directly to
996  \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
997  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
998  described  passed; these are described
999  .\" HTML <a href="#extradata">  .\" HTML <a href="#extradata">
1000  .\" </a>  .\" </a>
1001  below  below
1002  .\"  .\"
1003  in the section on matching a pattern.  in the section on matching a pattern.
1004  .P  .P
1005  If studying the pattern does not produce any additional information  If studying the pattern does not produce any useful information,
1006  \fBpcre_study()\fP returns NULL. In that circumstance, if the calling program  \fBpcre_study()\fP returns NULL by default. In that circumstance, if the
1007  wants to pass any of the other fields to \fBpcre_exec()\fP, it must set up its  calling program wants to pass any of the other fields to \fBpcre_exec()\fP or
1008  own \fBpcre_extra\fP block.  \fBpcre_dfa_exec()\fP, it must set up its own \fBpcre_extra\fP block. However,
1009  .P  if \fBpcre_study()\fP is called with the PCRE_STUDY_EXTRA_NEEDED option, it
1010  The second argument of \fBpcre_study()\fP contains option bits. At present, no  returns a \fBpcre_extra\fP block even if studying did not find any additional
1011  options are defined, and this argument should always be zero.  information. It may still return NULL, however, if an error occurs in
1012    \fBpcre_study()\fP.
1013    .P
1014    The second argument of \fBpcre_study()\fP contains option bits. There are three
1015    further options in addition to PCRE_STUDY_EXTRA_NEEDED:
1016    .sp
1017      PCRE_STUDY_JIT_COMPILE
1018      PCRE_STUDY_JIT_PARTIAL_HARD_COMPILE
1019      PCRE_STUDY_JIT_PARTIAL_SOFT_COMPILE
1020    .sp
1021    If any of these are set, and the just-in-time compiler is available, the
1022    pattern is further compiled into machine code that executes much faster than
1023    the \fBpcre_exec()\fP interpretive matching function. If the just-in-time
1024    compiler is not available, these options are ignored. All undefined bits in the
1025    \fIoptions\fP argument must be zero.
1026    .P
1027    JIT compilation is a heavyweight optimization. It can take some time for
1028    patterns to be analyzed, and for one-off matches and simple patterns the
1029    benefit of faster execution might be offset by a much slower study time.
1030    Not all patterns can be optimized by the JIT compiler. For those that cannot be
1031    handled, matching automatically falls back to the \fBpcre_exec()\fP
1032    interpreter. For more details, see the
1033    .\" HREF
1034    \fBpcrejit\fP
1035    .\"
1036    documentation.
1037  .P  .P
1038  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
1039  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 716  static string that is part of the librar Line 1042  static string that is part of the librar
1042  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
1043  sure that it has run successfully.  sure that it has run successfully.
1044  .P  .P
1045  This is a typical call to \fBpcre_study\fP():  When you are finished with a pattern, you can free the memory used for the
1046    study data by calling \fBpcre_free_study()\fP. This function was added to the
1047    API for release 8.20. For earlier versions, the memory could be freed with
1048    \fBpcre_free()\fP, just like the pattern itself. This will still work in cases
1049    where JIT optimization is not used, but it is advisable to change to the new
1050    function when convenient.
1051    .P
1052    This is a typical way in which \fBpcre_study\fP() is used (except that in a
1053    real application there should be tests for errors):
1054  .sp  .sp
1055    pcre_extra *pe;    int rc;
1056    pe = pcre_study(    pcre *re;
1057      pcre_extra *sd;
1058      re = pcre_compile("pattern", 0, &error, &erroroffset, NULL);
1059      sd = pcre_study(
1060      re,             /* result of pcre_compile() */      re,             /* result of pcre_compile() */
1061      0,              /* no options exist */      0,              /* no options */
1062      &error);        /* set to NULL or points to a message */      &error);        /* set to NULL or points to a message */
1063  .sp    rc = pcre_exec(   /* see below for details of pcre_exec() options */
1064  At present, studying a pattern is useful only for non-anchored patterns that do      re, sd, "subject", 7, 0, 0, ovector, 30);
1065  not have a single fixed starting character. A bitmap of possible starting    ...
1066  bytes is created.    pcre_free_study(sd);
1067      pcre_free(re);
1068    .sp
1069    Studying a pattern does two things: first, a lower bound for the length of
1070    subject string that is needed to match the pattern is computed. This does not
1071    mean that there are any strings of that length that match, but it does
1072    guarantee that no shorter strings match. The value is used to avoid wasting
1073    time by trying to match strings that are shorter than the lower bound. You can
1074    find out the value in a calling program via the \fBpcre_fullinfo()\fP function.
1075    .P
1076    Studying a pattern is also useful for non-anchored patterns that do not have a
1077    single fixed starting character. A bitmap of possible starting bytes is
1078    created. This speeds up finding a position in the subject at which to start
1079    matching. (In 16-bit mode, the bitmap is used for 16-bit values less than 256.
1080    In 32-bit mode, the bitmap is used for 32-bit values less than 256.)
1081    .P
1082    These two optimizations apply to both \fBpcre_exec()\fP and
1083    \fBpcre_dfa_exec()\fP, and the information is also used by the JIT compiler.
1084    The optimizations can be disabled by setting the PCRE_NO_START_OPTIMIZE option
1085    when calling \fBpcre_exec()\fP or \fBpcre_dfa_exec()\fP, but if this is done,
1086    JIT execution is also disabled. You might want to do this if your pattern
1087    contains callouts or (*MARK) and you want to make use of these facilities in
1088    cases where matching fails. See the discussion of PCRE_NO_START_OPTIMIZE
1089    .\" HTML <a href="#execoptions">
1090    .\" </a>
1091    below.
1092    .\"
1093  .  .
1094  .  .
1095  .\" HTML <a name="localesupport"></a>  .\" HTML <a name="localesupport"></a>
# Line 735  bytes is created. Line 1098  bytes is created.
1098  .sp  .sp
1099  PCRE handles caseless matching, and determines whether characters are letters,  PCRE handles caseless matching, and determines whether characters are letters,
1100  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
1101  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
1102  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
1103  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
1104  support. The use of locales with Unicode is discouraged. If you are handling  Unicode character property support. Alternatively, the PCRE_UCP option can be
1105  characters with codes greater than 128, you should either use UTF-8 and  set at compile time; this causes \ew and friends to use Unicode property
1106  Unicode, or use locales, but not try to mix the two.  support instead of built-in tables. The use of locales with Unicode is
1107    discouraged. If you are handling characters with codes greater than 128, you
1108    should either use UTF-8 and Unicode, or use locales, but not try to mix the
1109    two.
1110  .P  .P
1111  PCRE contains an internal set of tables that are used when the final argument  PCRE contains an internal set of tables that are used when the final argument
1112  of \fBpcre_compile()\fP is NULL. These are sufficient for many applications.  of \fBpcre_compile()\fP is NULL. These are sufficient for many applications.
# Line 785  one in which it was compiled. Passing ta Line 1151  one in which it was compiled. Passing ta
1151  below in the section on matching a pattern.  below in the section on matching a pattern.
1152  .  .
1153  .  .
1154    .\" HTML <a name="infoaboutpattern"></a>
1155  .SH "INFORMATION ABOUT A PATTERN"  .SH "INFORMATION ABOUT A PATTERN"
1156  .rs  .rs
1157  .sp  .sp
# Line 793  below in the section on matching a patte Line 1160  below in the section on matching a patte
1160  .B int \fIwhat\fP, void *\fIwhere\fP);  .B int \fIwhat\fP, void *\fIwhere\fP);
1161  .PP  .PP
1162  The \fBpcre_fullinfo()\fP function returns information about a compiled  The \fBpcre_fullinfo()\fP function returns information about a compiled
1163  pattern. It replaces the obsolete \fBpcre_info()\fP function, which is  pattern. It replaces the \fBpcre_info()\fP function, which was removed from the
1164  nevertheless retained for backwards compability (and is documented below).  library at version 8.30, after more than 10 years of obsolescence.
1165  .P  .P
1166  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
1167  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 803  information is required, and the fourth Line 1170  information is required, and the fourth
1170  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
1171  the following negative numbers:  the following negative numbers:
1172  .sp  .sp
1173    PCRE_ERROR_NULL       the argument \fIcode\fP was NULL    PCRE_ERROR_NULL           the argument \fIcode\fP was NULL
1174                          the argument \fIwhere\fP was NULL                              the argument \fIwhere\fP was NULL
1175    PCRE_ERROR_BADMAGIC   the "magic number" was not found    PCRE_ERROR_BADMAGIC       the "magic number" was not found
1176    PCRE_ERROR_BADOPTION  the value of \fIwhat\fP was invalid    PCRE_ERROR_BADENDIANNESS  the pattern was compiled with different
1177                                endianness
1178      PCRE_ERROR_BADOPTION      the value of \fIwhat\fP was invalid
1179  .sp  .sp
1180  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
1181  check against passing an arbitrary memory pointer. Here is a typical call of  check against passing an arbitrary memory pointer. The endianness error can
1182  \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
1183    a typical call of \fBpcre_fullinfo()\fP, to obtain the length of the compiled
1184    pattern:
1185  .sp  .sp
1186    int rc;    int rc;
1187    size_t length;    size_t length;
1188    rc = pcre_fullinfo(    rc = pcre_fullinfo(
1189      re,               /* result of pcre_compile() */      re,               /* result of pcre_compile() */
1190      pe,               /* result of pcre_study(), or NULL */      sd,               /* result of pcre_study(), or NULL */
1191      PCRE_INFO_SIZE,   /* what is required */      PCRE_INFO_SIZE,   /* what is required */
1192      &length);         /* where to put the data */      &length);         /* where to put the data */
1193  .sp  .sp
# Line 844  a NULL table pointer. Line 1215  a NULL table pointer.
1215  .sp  .sp
1216    PCRE_INFO_FIRSTBYTE    PCRE_INFO_FIRSTBYTE
1217  .sp  .sp
1218  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
1219  non-anchored pattern. The fourth argument should point to an \fBint\fP  non-anchored pattern. (The name of this option refers to the 8-bit library,
1220  variable. (This option used to be called PCRE_INFO_FIRSTCHAR; the old name is  where data units are bytes.) The fourth argument should point to an \fBint\fP
1221  still recognized for backwards compatibility.)  variable.
1222  .P  .P
1223  If there is a fixed first byte, for example, from a pattern such as  If there is a fixed first value, for example, the letter "c" from a pattern
1224  (cat|cow|coyote), its value is returned. Otherwise, if either  such as (cat|cow|coyote), its value is returned. In the 8-bit library, the
1225    value is always less than 256. In the 16-bit library the value can be up to
1226    0xffff. In the 32-bit library the value can be up to 0x10ffff.
1227    .P
1228    If there is no fixed first value, and if either
1229  .sp  .sp
1230  (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
1231  starts with "^", or  starts with "^", or
# Line 861  starts with "^", or Line 1236  starts with "^", or
1236  -1 is returned, indicating that the pattern matches only at the start of a  -1 is returned, indicating that the pattern matches only at the start of a
1237  subject string or after any newline within the string. Otherwise -2 is  subject string or after any newline within the string. Otherwise -2 is
1238  returned. For anchored patterns, -2 is returned.  returned. For anchored patterns, -2 is returned.
1239    .P
1240    Since for the 32-bit library using the non-UTF-32 mode, this function is unable
1241    to return the full 32-bit range of the character, this value is deprecated;
1242    instead the PCRE_INFO_FIRSTCHARACTERFLAGS and PCRE_INFO_FIRSTCHARACTER values
1243    should be used.
1244  .sp  .sp
1245    PCRE_INFO_FIRSTTABLE    PCRE_INFO_FIRSTTABLE
1246  .sp  .sp
1247  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
1248  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
1249  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
1250  fourth argument should point to an \fBunsigned char *\fP variable.  fourth argument should point to an \fBunsigned char *\fP variable.
1251  .sp  .sp
1252      PCRE_INFO_HASCRORLF
1253    .sp
1254    Return 1 if the pattern contains any explicit matches for CR or LF characters,
1255    otherwise 0. The fourth argument should point to an \fBint\fP variable. An
1256    explicit match is either a literal CR or LF character, or \er or \en.
1257    .sp
1258      PCRE_INFO_JCHANGED
1259    .sp
1260    Return 1 if the (?J) or (?-J) option setting is used in the pattern, otherwise
1261    0. The fourth argument should point to an \fBint\fP variable. (?J) and
1262    (?-J) set and unset the local PCRE_DUPNAMES option, respectively.
1263    .sp
1264      PCRE_INFO_JIT
1265    .sp
1266    Return 1 if the pattern was studied with one of the JIT options, and
1267    just-in-time compiling was successful. The fourth argument should point to an
1268    \fBint\fP variable. A return value of 0 means that JIT support is not available
1269    in this version of PCRE, or that the pattern was not studied with a JIT option,
1270    or that the JIT compiler could not handle this particular pattern. See the
1271    .\" HREF
1272    \fBpcrejit\fP
1273    .\"
1274    documentation for details of what can and cannot be handled.
1275    .sp
1276      PCRE_INFO_JITSIZE
1277    .sp
1278    If the pattern was successfully studied with a JIT option, return the size of
1279    the JIT compiled code, otherwise return zero. The fourth argument should point
1280    to a \fBsize_t\fP variable.
1281    .sp
1282    PCRE_INFO_LASTLITERAL    PCRE_INFO_LASTLITERAL
1283  .sp  .sp
1284  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
1285  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
1286  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
1287  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
1288  follows something of variable length. For example, for the pattern  only if it follows something of variable length. For example, for the pattern
1289  /^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
1290  is -1.  is -1.
1291    .P
1292    Since for the 32-bit library using the non-UTF-32 mode, this function is unable
1293    to return the full 32-bit range of the character, this value is deprecated;
1294    instead the PCRE_INFO_REQUIREDCHARFLAGS and PCRE_INFO_REQUIREDCHAR values should
1295    be used.
1296    .sp
1297      PCRE_INFO_MAXLOOKBEHIND
1298    .sp
1299    Return the number of characters (NB not bytes) in the longest lookbehind
1300    assertion in the pattern. Note that the simple assertions \eb and \eB require a
1301    one-character lookbehind. This information is useful when doing multi-segment
1302    matching using the partial matching facilities.
1303    .sp
1304      PCRE_INFO_MINLENGTH
1305    .sp
1306    If the pattern was studied and a minimum length for matching subject strings
1307    was computed, its value is returned. Otherwise the returned value is -1. The
1308    value is a number of characters, which in UTF-8 mode may be different from the
1309    number of bytes. The fourth argument should point to an \fBint\fP variable. A
1310    non-negative value is a lower bound to the length of any matching string. There
1311    may not be any strings of that length that do actually match, but every string
1312    that does match is at least that long.
1313  .sp  .sp
1314    PCRE_INFO_NAMECOUNT    PCRE_INFO_NAMECOUNT
1315    PCRE_INFO_NAMEENTRYSIZE    PCRE_INFO_NAMEENTRYSIZE
# Line 897  The map consists of a number of fixed-si Line 1329  The map consists of a number of fixed-si
1329  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
1330  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
1331  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
1332  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
1333  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,
1334  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
1335  alphabetical order. When PCRE_DUPNAMES is set, duplicate names are in order of  16-bit data units, the first of which contains the parenthesis number.
1336  their parentheses numbers. For example, consider the following pattern (assume  In the 32-bit library, the pointer points to 32-bit data units, the first of
1337  PCRE_EXTENDED is set, so white space - including newlines - is ignored):  which contains the parenthesis number. The rest
1338    of the entry is the corresponding name, zero terminated.
1339    .P
1340    The names are in alphabetical order. Duplicate names may appear if (?| is used
1341    to create multiple groups with the same number, as described in the
1342    .\" HTML <a href="pcrepattern.html#dupsubpatternnumber">
1343    .\" </a>
1344    section on duplicate subpattern numbers
1345    .\"
1346    in the
1347    .\" HREF
1348    \fBpcrepattern\fP
1349    .\"
1350    page. Duplicate names for subpatterns with different numbers are permitted only
1351    if PCRE_DUPNAMES is set. In all cases of duplicate names, they appear in the
1352    table in the order in which they were found in the pattern. In the absence of
1353    (?| this is the order of increasing number; when (?| is used this is not
1354    necessarily the case because later subpatterns may have lower numbers.
1355    .P
1356    As a simple example of the name/number table, consider the following pattern
1357    after compilation by the 8-bit library (assume PCRE_EXTENDED is set, so white
1358    space - including newlines - is ignored):
1359  .sp  .sp
1360  .\" JOIN  .\" JOIN
1361    (?<date> (?<year>(\ed\ed)?\ed\ed) -    (?<date> (?<year>(\ed\ed)?\ed\ed) -
# Line 921  When writing code to extract data from n Line 1374  When writing code to extract data from n
1374  name-to-number map, remember that the length of the entries is likely to be  name-to-number map, remember that the length of the entries is likely to be
1375  different for each compiled pattern.  different for each compiled pattern.
1376  .sp  .sp
1377      PCRE_INFO_OKPARTIAL
1378    .sp
1379    Return 1 if the pattern can be used for partial matching with
1380    \fBpcre_exec()\fP, otherwise 0. The fourth argument should point to an
1381    \fBint\fP variable. From release 8.00, this always returns 1, because the
1382    restrictions that previously applied to partial matching have been lifted. The
1383    .\" HREF
1384    \fBpcrepartial\fP
1385    .\"
1386    documentation gives details of partial matching.
1387    .sp
1388    PCRE_INFO_OPTIONS    PCRE_INFO_OPTIONS
1389  .sp  .sp
1390  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
1391  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
1392  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
1393  top-level option settings within the pattern itself.  top-level option settings at the start of the pattern itself. In other words,
1394    they are the options that will be in force when matching starts. For example,
1395    if the pattern /(?im)abc(?-i)d/ is compiled with the PCRE_EXTENDED option, the
1396    result is PCRE_CASELESS, PCRE_MULTILINE, and PCRE_EXTENDED.
1397  .P  .P
1398  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
1399  alternatives begin with one of the following:  alternatives begin with one of the following:
# Line 943  For such patterns, the PCRE_ANCHORED bit Line 1410  For such patterns, the PCRE_ANCHORED bit
1410  .sp  .sp
1411    PCRE_INFO_SIZE    PCRE_INFO_SIZE
1412  .sp  .sp
1413  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
1414  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
1415  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
1416  variable.  \fBpcre_compile()\fP. The value that is passed as the argument to
1417    \fBpcre_malloc()\fP when \fBpcre_compile()\fP is getting memory in which to
1418    place the compiled data is the value returned by this option plus the size of
1419    the \fBpcre\fP structure. Studying a compiled pattern, with or without JIT,
1420    does not alter the value returned by this option.
1421  .sp  .sp
1422    PCRE_INFO_STUDYSIZE    PCRE_INFO_STUDYSIZE
1423  .sp  .sp
1424  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
1425  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
1426  \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
1427  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
1428  \fBsize_t\fP variable.  to record information that will speed up matching (see the section entitled
1429  .  .\" HTML <a href="#studyingapattern">
1430  .  .\" </a>
1431  .SH "OBSOLETE INFO FUNCTION"  "Studying a pattern"
1432  .rs  .\"
1433    above). The format of the \fIstudy_data\fP block is private, but its length
1434    is made available via this option so that it can be saved and restored (see the
1435    .\" HREF
1436    \fBpcreprecompile\fP
1437    .\"
1438    documentation for details).
1439  .sp  .sp
1440  .B int pcre_info(const pcre *\fIcode\fP, int *\fIoptptr\fP, int    PCRE_INFO_FIRSTCHARACTERFLAGS
1441  .B *\fIfirstcharptr\fP);  .sp
1442  .PP  Return information about the first data unit of any matched string, for a
1443  The \fBpcre_info()\fP function is now obsolete because its interface is too  non-anchored pattern. The fourth argument should point to an \fBint\fP
1444  restrictive to return all the available data about a compiled pattern. New  variable.
1445  programs should use \fBpcre_fullinfo()\fP instead. The yield of  .P
1446  \fBpcre_info()\fP is the number of capturing subpatterns, or one of the  If there is a fixed first value, for example, the letter "c" from a pattern
1447  following negative numbers:  such as (cat|cow|coyote), 1 is returned, and the character value can be
1448    retrieved using PCRE_INFO_FIRSTCHARACTER.
1449    .P
1450    If there is no fixed first value, and if either
1451    .sp
1452    (a) the pattern was compiled with the PCRE_MULTILINE option, and every branch
1453    starts with "^", or
1454    .sp
1455    (b) every branch of the pattern starts with ".*" and PCRE_DOTALL is not set
1456    (if it were set, the pattern would be anchored),
1457    .sp
1458    2 is returned, indicating that the pattern matches only at the start of a
1459    subject string or after any newline within the string. Otherwise 0 is
1460    returned. For anchored patterns, 0 is returned.
1461  .sp  .sp
1462    PCRE_ERROR_NULL       the argument \fIcode\fP was NULL    PCRE_INFO_FIRSTCHARACTER
   PCRE_ERROR_BADMAGIC   the "magic number" was not found  
1463  .sp  .sp
1464  If the \fIoptptr\fP argument is not NULL, a copy of the options with which the  Return the fixed first character value, if PCRE_INFO_FIRSTCHARACTERFLAGS
1465  pattern was compiled is placed in the integer it points to (see  returned 1; otherwise returns 0. The fourth argument should point to an
1466  PCRE_INFO_OPTIONS above).  \fBuint_t\fP variable.
1467  .P  .P
1468  If the pattern is not anchored and the \fIfirstcharptr\fP argument is not NULL,  In the 8-bit library, the value is always less than 256. In the 16-bit library
1469  it is used to pass back information about the first character of any matched  the value can be up to 0xffff. In the 32-bit library in UTF-32 mode the value
1470  string (see PCRE_INFO_FIRSTBYTE above).  can be up to 0x10ffff, and up to 0xffffffff when not using UTF-32 mode.
1471    .P
1472    If there is no fixed first value, and if either
1473    .sp
1474    (a) the pattern was compiled with the PCRE_MULTILINE option, and every branch
1475    starts with "^", or
1476    .sp
1477    (b) every branch of the pattern starts with ".*" and PCRE_DOTALL is not set
1478    (if it were set, the pattern would be anchored),
1479    .sp
1480    -1 is returned, indicating that the pattern matches only at the start of a
1481    subject string or after any newline within the string. Otherwise -2 is
1482    returned. For anchored patterns, -2 is returned.
1483    .sp
1484      PCRE_INFO_REQUIREDCHARFLAGS
1485    .sp
1486    Returns 1 if there is a rightmost literal data unit that must exist in any
1487    matched string, other than at its start. The fourth argument should  point to
1488    an \fBint\fP variable. If there is no such value, 0 is returned. If returning
1489    1, the character value itself can be retrieved using PCRE_INFO_REQUIREDCHAR.
1490    .P
1491    For anchored patterns, a last literal value is recorded only if it follows
1492    something of variable length. For example, for the pattern /^a\ed+z\ed+/ the
1493    returned value 1 (with "z" returned from PCRE_INFO_REQUIREDCHAR), but for
1494    /^a\edz\ed/ the returned value is 0.
1495    .sp
1496      PCRE_INFO_REQUIREDCHAR
1497    .sp
1498    Return the value of the rightmost literal data unit that must exist in any
1499    matched string, other than at its start, if such a value has been recorded. The
1500    fourth argument should point to an \fBuint32_t\fP variable. If there is no such
1501    value, 0 is returned.
1502  .  .
1503  .  .
1504  .SH "REFERENCE COUNTS"  .SH "REFERENCE COUNTS"
# Line 1015  is different. (This seems a highly unlik Line 1535  is different. (This seems a highly unlik
1535  .P  .P
1536  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
1537  compiled pattern, which is passed in the \fIcode\fP argument. If the  compiled pattern, which is passed in the \fIcode\fP argument. If the
1538  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
1539  \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
1540  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
1541  also an alternative matching function, which is described  different subject strings with the same pattern.
1542    .P
1543    This function is the main matching facility of the library, and it operates in
1544    a Perl-like manner. For specialist use there is also an alternative matching
1545    function, which is described
1546  .\" HTML <a href="#dfamatch">  .\" HTML <a href="#dfamatch">
1547  .\" </a>  .\" </a>
1548  below  below
# Line 1049  Here is an example of a simple call to \ Line 1573  Here is an example of a simple call to \
1573      ovector,        /* vector of integers for substring information */      ovector,        /* vector of integers for substring information */
1574      30);            /* number of elements (NOT size in bytes) */      30);            /* number of elements (NOT size in bytes) */
1575  .  .
1576    .
1577  .\" HTML <a name="extradata"></a>  .\" HTML <a name="extradata"></a>
1578  .SS "Extra data for \fBpcre_exec()\fR"  .SS "Extra data for \fBpcre_exec()\fR"
1579  .rs  .rs
# Line 1061  fields (not necessarily in this order): Line 1586  fields (not necessarily in this order):
1586  .sp  .sp
1587    unsigned long int \fIflags\fP;    unsigned long int \fIflags\fP;
1588    void *\fIstudy_data\fP;    void *\fIstudy_data\fP;
1589      void *\fIexecutable_jit\fP;
1590    unsigned long int \fImatch_limit\fP;    unsigned long int \fImatch_limit\fP;
1591    unsigned long int \fImatch_limit_recursion\fP;    unsigned long int \fImatch_limit_recursion\fP;
1592    void *\fIcallout_data\fP;    void *\fIcallout_data\fP;
1593    const unsigned char *\fItables\fP;    const unsigned char *\fItables\fP;
1594      unsigned char **\fImark\fP;
1595  .sp  .sp
1596  The \fIflags\fP field is a bitmap that specifies which of the other fields  In the 16-bit version of this structure, the \fImark\fP field has type
1597  are set. The flag bits are:  "PCRE_UCHAR16 **".
1598  .sp  .sp
1599    PCRE_EXTRA_STUDY_DATA  In the 32-bit version of this structure, the \fImark\fP field has type
1600    "PCRE_UCHAR32 **".
1601    .P
1602    The \fIflags\fP field is used to specify which of the other fields are set. The
1603    flag bits are:
1604    .sp
1605      PCRE_EXTRA_CALLOUT_DATA
1606      PCRE_EXTRA_EXECUTABLE_JIT
1607      PCRE_EXTRA_MARK
1608    PCRE_EXTRA_MATCH_LIMIT    PCRE_EXTRA_MATCH_LIMIT
1609    PCRE_EXTRA_MATCH_LIMIT_RECURSION    PCRE_EXTRA_MATCH_LIMIT_RECURSION
1610    PCRE_EXTRA_CALLOUT_DATA    PCRE_EXTRA_STUDY_DATA
1611    PCRE_EXTRA_TABLES    PCRE_EXTRA_TABLES
1612  .sp  .sp
1613  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
1614  \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
1615  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
1616  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 other
1617    fields and their corresponding flag bits.
1618  .P  .P
1619  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
1620  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,
1621  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
1622  classic example is the use of nested unlimited repeats.  classic example is a pattern that uses nested unlimited repeats.
1623  .P  .P
1624  Internally, PCRE uses a function called \fBmatch()\fP which it calls repeatedly  Internally, \fBpcre_exec()\fP uses a function called \fBmatch()\fP, which it
1625  (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
1626  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
1627  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
1628  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
1629  string.  in the subject string.
1630    .P
1631    When \fBpcre_exec()\fP is called with a pattern that was successfully studied
1632    with a JIT option, the way that the matching is executed is entirely different.
1633    However, there is still the possibility of runaway matching that goes on for a
1634    very long time, and so the \fImatch_limit\fP value is also used in this case
1635    (but in a different way) to limit how long the matching can continue.
1636  .P  .P
1637  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
1638  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 1105  limits the depth of recursion. The recur Line 1647  limits the depth of recursion. The recur
1647  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.
1648  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.
1649  .P  .P
1650  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
1651  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
1652  amount of heap memory that can be used.  stack, the amount of heap memory that can be used. This limit is not relevant,
1653    and is ignored, when matching is done using JIT compiled code.
1654  .P  .P
1655  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
1656  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 1116  with a \fBpcre_extra\fP block in which \ Line 1659  with a \fBpcre_extra\fP block in which \
1659  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
1660  is exceeded, \fBpcre_exec()\fP returns PCRE_ERROR_RECURSIONLIMIT.  is exceeded, \fBpcre_exec()\fP returns PCRE_ERROR_RECURSIONLIMIT.
1661  .P  .P
1662  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,
1663  which is described in the  and is described in the
1664  .\" HREF  .\" HREF
1665  \fBpcrecallout\fP  \fBpcrecallout\fP
1666  .\"  .\"
# Line 1136  called. See the Line 1679  called. See the
1679  \fBpcreprecompile\fP  \fBpcreprecompile\fP
1680  .\"  .\"
1681  documentation for a discussion of saving compiled patterns for later use.  documentation for a discussion of saving compiled patterns for later use.
1682    .P
1683    If PCRE_EXTRA_MARK is set in the \fIflags\fP field, the \fImark\fP field must
1684    be set to point to a suitable variable. If the pattern contains any
1685    backtracking control verbs such as (*MARK:NAME), and the execution ends up with
1686    a name to pass back, a pointer to the name string (zero terminated) is placed
1687    in the variable pointed to by the \fImark\fP field. The names are within the
1688    compiled pattern; if you wish to retain such a name you must copy it before
1689    freeing the memory of a compiled pattern. If there is no name to pass back, the
1690    variable pointed to by the \fImark\fP field is set to NULL. For details of the
1691    backtracking control verbs, see the section entitled
1692    .\" HTML <a href="pcrepattern#backtrackcontrol">
1693    .\" </a>
1694    "Backtracking control"
1695    .\"
1696    in the
1697    .\" HREF
1698    \fBpcrepattern\fP
1699    .\"
1700    documentation.
1701    .
1702  .  .
1703    .\" HTML <a name="execoptions"></a>
1704  .SS "Option bits for \fBpcre_exec()\fP"  .SS "Option bits for \fBpcre_exec()\fP"
1705  .rs  .rs
1706  .sp  .sp
1707  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
1708  zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_\fIxxx\fP,  zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_\fIxxx\fP,
1709  PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NO_UTF8_CHECK and PCRE_PARTIAL.  PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NOTEMPTY_ATSTART,
1710    PCRE_NO_START_OPTIMIZE, PCRE_NO_UTF8_CHECK, PCRE_PARTIAL_HARD, and
1711    PCRE_PARTIAL_SOFT.
1712    .P
1713    If the pattern was successfully studied with one of the just-in-time (JIT)
1714    compile options, the only supported options for JIT execution are
1715    PCRE_NO_UTF8_CHECK, PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY,
1716    PCRE_NOTEMPTY_ATSTART, PCRE_PARTIAL_HARD, and PCRE_PARTIAL_SOFT. If an
1717    unsupported option is used, JIT execution is disabled and the normal
1718    interpretive code in \fBpcre_exec()\fP is run.
1719  .sp  .sp
1720    PCRE_ANCHORED    PCRE_ANCHORED
1721  .sp  .sp
# Line 1151  matching position. If a pattern was comp Line 1724  matching position. If a pattern was comp
1724  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
1725  matching time.  matching time.
1726  .sp  .sp
1727      PCRE_BSR_ANYCRLF
1728      PCRE_BSR_UNICODE
1729    .sp
1730    These options (which are mutually exclusive) control what the \eR escape
1731    sequence matches. The choice is either to match only CR, LF, or CRLF, or to
1732    match any Unicode newline sequence. These options override the choice that was
1733    made or defaulted when the pattern was compiled.
1734    .sp
1735    PCRE_NEWLINE_CR    PCRE_NEWLINE_CR
1736    PCRE_NEWLINE_LF    PCRE_NEWLINE_LF
1737    PCRE_NEWLINE_CRLF    PCRE_NEWLINE_CRLF
# Line 1162  the pattern was compiled. For details, s Line 1743  the pattern was compiled. For details, s
1743  \fBpcre_compile()\fP above. During matching, the newline choice affects the  \fBpcre_compile()\fP above. During matching, the newline choice affects the
1744  behaviour of the dot, circumflex, and dollar metacharacters. It may also alter  behaviour of the dot, circumflex, and dollar metacharacters. It may also alter
1745  the way the match position is advanced after a match failure for an unanchored  the way the match position is advanced after a match failure for an unanchored
1746  pattern. When PCRE_NEWLINE_CRLF, PCRE_NEWLINE_ANYCRLF, or PCRE_NEWLINE_ANY is  pattern.
1747  set, and a match attempt fails when the current position is at a CRLF sequence,  .P
1748  the match position is advanced by two characters instead of one, in other  When PCRE_NEWLINE_CRLF, PCRE_NEWLINE_ANYCRLF, or PCRE_NEWLINE_ANY is set, and a
1749  words, to after the CRLF.  match attempt for an unanchored pattern fails when the current position is at a
1750    CRLF sequence, and the pattern contains no explicit matches for CR or LF
1751    characters, the match position is advanced by two characters instead of one, in
1752    other words, to after the CRLF.
1753    .P
1754    The above rule is a compromise that makes the most common cases work as
1755    expected. For example, if the pattern is .+A (and the PCRE_DOTALL option is not
1756    set), it does not match the string "\er\enA" because, after failing at the
1757    start, it skips both the CR and the LF before retrying. However, the pattern
1758    [\er\en]A does match that string, because it contains an explicit CR or LF
1759    reference, and so advances only by one character after the first failure.
1760    .P
1761    An explicit match for CR of LF is either a literal appearance of one of those
1762    characters, or one of the \er or \en escape sequences. Implicit matches such as
1763    [^X] do not count, nor does \es (which includes CR and LF in the characters
1764    that it matches).
1765    .P
1766    Notwithstanding the above, anomalous effects may still occur when CRLF is a
1767    valid newline sequence and explicit \er or \en escapes appear in the pattern.
1768  .sp  .sp
1769    PCRE_NOTBOL    PCRE_NOTBOL
1770  .sp  .sp
# Line 1191  match the empty string, the entire match Line 1790  match the empty string, the entire match
1790  .sp  .sp
1791    a?b?    a?b?
1792  .sp  .sp
1793  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
1794  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
1795  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".
1796  .P  .sp
1797  Perl has no direct equivalent of PCRE_NOTEMPTY, but it does make a special case    PCRE_NOTEMPTY_ATSTART
1798  of a pattern match of the empty string within its \fBsplit()\fP function, and  .sp
1799  when using the /g modifier. It is possible to emulate Perl's behaviour after  This is like PCRE_NOTEMPTY, except that an empty string match that is not at
1800  matching a null string by first trying the match again at the same offset with  the start of the subject is permitted. If the pattern is anchored, such a match
1801  PCRE_NOTEMPTY and PCRE_ANCHORED, and then if that fails by advancing the  can occur only if the pattern contains \eK.
1802  starting offset (see below) and trying an ordinary match again. There is some  .P
1803  code that demonstrates how to do this in the \fIpcredemo.c\fP sample program.  Perl has no direct equivalent of PCRE_NOTEMPTY or PCRE_NOTEMPTY_ATSTART, but it
1804    does make a special case of a pattern match of the empty string within its
1805    \fBsplit()\fP function, and when using the /g modifier. It is possible to
1806    emulate Perl's behaviour after matching a null string by first trying the match
1807    again at the same offset with PCRE_NOTEMPTY_ATSTART and PCRE_ANCHORED, and then
1808    if that fails, by advancing the starting offset (see below) and trying an
1809    ordinary match again. There is some code that demonstrates how to do this in
1810    the
1811    .\" HREF
1812    \fBpcredemo\fP
1813    .\"
1814    sample program. In the most general case, you have to check to see if the
1815    newline convention recognizes CRLF as a newline, and if so, and the current
1816    character is CR followed by LF, advance the starting offset by two characters
1817    instead of one.
1818    .sp
1819      PCRE_NO_START_OPTIMIZE
1820    .sp
1821    There are a number of optimizations that \fBpcre_exec()\fP uses at the start of
1822    a match, in order to speed up the process. For example, if it is known that an
1823    unanchored match must start with a specific character, it searches the subject
1824    for that character, and fails immediately if it cannot find it, without
1825    actually running the main matching function. This means that a special item
1826    such as (*COMMIT) at the start of a pattern is not considered until after a
1827    suitable starting point for the match has been found. When callouts or (*MARK)
1828    items are in use, these "start-up" optimizations can cause them to be skipped
1829    if the pattern is never actually used. The start-up optimizations are in effect
1830    a pre-scan of the subject that takes place before the pattern is run.
1831    .P
1832    The PCRE_NO_START_OPTIMIZE option disables the start-up optimizations, possibly
1833    causing performance to suffer, but ensuring that in cases where the result is
1834    "no match", the callouts do occur, and that items such as (*COMMIT) and (*MARK)
1835    are considered at every possible starting position in the subject string. If
1836    PCRE_NO_START_OPTIMIZE is set at compile time, it cannot be unset at matching
1837    time. The use of PCRE_NO_START_OPTIMIZE disables JIT execution; when it is set,
1838    matching is always done using interpretively.
1839    .P
1840    Setting PCRE_NO_START_OPTIMIZE can change the outcome of a matching operation.
1841    Consider the pattern
1842    .sp
1843      (*COMMIT)ABC
1844    .sp
1845    When this is compiled, PCRE records the fact that a match must start with the
1846    character "A". Suppose the subject string is "DEFABC". The start-up
1847    optimization scans along the subject, finds "A" and runs the first match
1848    attempt from there. The (*COMMIT) item means that the pattern must match the
1849    current starting position, which in this case, it does. However, if the same
1850    match is run with PCRE_NO_START_OPTIMIZE set, the initial scan along the
1851    subject string does not happen. The first match attempt is run starting from
1852    "D" and when this fails, (*COMMIT) prevents any further matches being tried, so
1853    the overall result is "no match". If the pattern is studied, more start-up
1854    optimizations may be used. For example, a minimum length for the subject may be
1855    recorded. Consider the pattern
1856    .sp
1857      (*MARK:A)(X|Y)
1858    .sp
1859    The minimum length for a match is one character. If the subject is "ABC", there
1860    will be attempts to match "ABC", "BC", "C", and then finally an empty string.
1861    If the pattern is studied, the final attempt does not take place, because PCRE
1862    knows that the subject is too short, and so the (*MARK) is never encountered.
1863    In this case, studying the pattern does not affect the overall match result,
1864    which is still "no match", but it does affect the auxiliary information that is
1865    returned.
1866  .sp  .sp
1867    PCRE_NO_UTF8_CHECK    PCRE_NO_UTF8_CHECK
1868  .sp  .sp
1869  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
1870  string is automatically checked when \fBpcre_exec()\fP is subsequently called.  string is automatically checked when \fBpcre_exec()\fP is subsequently called.
1871  The value of \fIstartoffset\fP is also checked to ensure that it points to the  The entire string is checked before any other processing takes place. The value
1872  start of a UTF-8 character. If an invalid UTF-8 sequence of bytes is found,  of \fIstartoffset\fP is also checked to ensure that it points to the start of a
1873  \fBpcre_exec()\fP returns the error PCRE_ERROR_BADUTF8. If \fIstartoffset\fP  UTF-8 character. There is a discussion about the
1874  contains an invalid value, PCRE_ERROR_BADUTF8_OFFSET is returned.  .\" HTML <a href="pcreunicode.html#utf8strings">
1875    .\" </a>
1876    validity of UTF-8 strings
1877    .\"
1878    in the
1879    .\" HREF
1880    \fBpcreunicode\fP
1881    .\"
1882    page. If an invalid sequence of bytes is found, \fBpcre_exec()\fP returns the
1883    error PCRE_ERROR_BADUTF8 or, if PCRE_PARTIAL_HARD is set and the problem is a
1884    truncated character at the end of the subject, PCRE_ERROR_SHORTUTF8. In both
1885    cases, information about the precise nature of the error may also be returned
1886    (see the descriptions of these errors in the section entitled \fIError return
1887    values from\fP \fBpcre_exec()\fP
1888    .\" HTML <a href="#errorlist">
1889    .\" </a>
1890    below).
1891    .\"
1892    If \fIstartoffset\fP contains a value that does not point to the start of a
1893    UTF-8 character (or to the end of the subject), PCRE_ERROR_BADUTF8_OFFSET is
1894    returned.
1895  .P  .P
1896  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
1897  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
1898  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
1899  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
1900  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
1901  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
1902  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
1903  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
1904  UTF-8 character, is undefined. Your program may crash.  undefined. Your program may crash.
1905  .sp  .sp
1906    PCRE_PARTIAL    PCRE_PARTIAL_HARD
1907  .sp    PCRE_PARTIAL_SOFT
1908  This option turns on the partial matching feature. If the subject string fails  .sp
1909  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
1910  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
1911  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
1912  characters), \fBpcre_exec()\fP returns PCRE_ERROR_PARTIAL instead of  not enough subject characters to complete the match. If this happens when
1913  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
1914  may appear in the pattern. These are discussed in the  testing any remaining alternatives. Only if no complete match can be found is
1915    PCRE_ERROR_PARTIAL returned instead of PCRE_ERROR_NOMATCH. In other words,
1916    PCRE_PARTIAL_SOFT says that the caller is prepared to handle a partial match,
1917    but only if no complete match can be found.
1918    .P
1919    If PCRE_PARTIAL_HARD is set, it overrides PCRE_PARTIAL_SOFT. In this case, if a
1920    partial match is found, \fBpcre_exec()\fP immediately returns
1921    PCRE_ERROR_PARTIAL, without considering any other alternatives. In other words,
1922    when PCRE_PARTIAL_HARD is set, a partial match is considered to be more
1923    important that an alternative complete match.
1924    .P
1925    In both cases, the portion of the string that was inspected when the partial
1926    match was found is set as the first matching string. There is a more detailed
1927    discussion of partial and multi-segment matching, with examples, in the
1928  .\" HREF  .\" HREF
1929  \fBpcrepartial\fP  \fBpcrepartial\fP
1930  .\"  .\"
1931  documentation.  documentation.
1932  .  .
1933    .
1934  .SS "The string to be matched by \fBpcre_exec()\fP"  .SS "The string to be matched by \fBpcre_exec()\fP"
1935  .rs  .rs
1936  .sp  .sp
1937  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
1938  \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
1939  \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
1940  UTF-8 character. Unlike the pattern string, the subject may contain binary zero  subject, \fBpcre_exec()\fP returns PCRE_ERROR_BADOFFSET. When the starting
1941  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,
1942  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
1943    point to the start of a UTF-8 character (or the end of the subject). Unlike the
1944    pattern string, the subject may contain binary zero bytes.
1945  .P  .P
1946  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
1947  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 1264  start of the subject, which is deemed to Line 1961  start of the subject, which is deemed to
1961  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
1962  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.
1963  .P  .P
1964    Finding all the matches in a subject is tricky when the pattern can match an
1965    empty string. It is possible to emulate Perl's /g behaviour by first trying the
1966    match again at the same offset, with the PCRE_NOTEMPTY_ATSTART and
1967    PCRE_ANCHORED options, and then if that fails, advancing the starting offset
1968    and trying an ordinary match again. There is some code that demonstrates how to
1969    do this in the
1970    .\" HREF
1971    \fBpcredemo\fP
1972    .\"
1973    sample program. In the most general case, you have to check to see if the
1974    newline convention recognizes CRLF as a newline, and if so, and the current
1975    character is CR followed by LF, advance the starting offset by two characters
1976    instead of one.
1977    .P
1978  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
1979  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
1980  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.
1981  .  .
1982    .
1983  .SS "How \fBpcre_exec()\fP returns captured substrings"  .SS "How \fBpcre_exec()\fP returns captured substrings"
1984  .rs  .rs
1985  .sp  .sp
# Line 1278  pattern. Following the usage in Jeffrey Line 1990  pattern. Following the usage in Jeffrey
1990  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
1991  kinds of parenthesized subpattern that do not cause substrings to be captured.  kinds of parenthesized subpattern that do not cause substrings to be captured.
1992  .P  .P
1993  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
1994  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
1995  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
1996  this argument is NOT the size of \fIovector\fP in bytes.  argument is NOT the size of \fIovector\fP in bytes.
1997  .P  .P
1998  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,
1999  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
2000  used as workspace by \fBpcre_exec()\fP while matching capturing subpatterns,  used as workspace by \fBpcre_exec()\fP while matching capturing subpatterns,
2001  and is not available for passing back information. The length passed in  and is not available for passing back information. The number passed in
2002  \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
2003  rounded down.  rounded down.
2004  .P  .P
2005  When a match is successful, information about captured substrings is returned  When a match is successful, information about captured substrings is returned
2006  in pairs of integers, starting at the beginning of \fIovector\fP, and  in pairs of integers, starting at the beginning of \fIovector\fP, and
2007  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
2008  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
2009  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
2010  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
2011  subject string matched by the entire pattern. The next pair is used for the  mode. They are not character counts.
2012  first capturing subpattern, and so on. The value returned by \fBpcre_exec()\fP  .P
2013  is one more than the highest numbered pair that has been set. For example, if  The first pair of integers, \fIovector[0]\fP and \fIovector[1]\fP, identify the
2014  two substrings have been captured, the returned value is 3. If there are no  portion of the subject string matched by the entire pattern. The next pair is
2015  capturing subpatterns, the return value from a successful match is 1,  used for the first capturing subpattern, and so on. The value returned by
2016  indicating that just the first pair of offsets has been set.  \fBpcre_exec()\fP is one more than the highest numbered pair that has been set.
2017    For example, if two substrings have been captured, the returned value is 3. If
2018    there are no capturing subpatterns, the return value from a successful match is
2019    1, indicating that just the first pair of offsets has been set.
2020  .P  .P
2021  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
2022  string that it matched that is returned.  string that it matched that is returned.
2023  .P  .P
2024  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
2025  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
2026  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 nor any captured
2027  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
2028  \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
2029  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
2030  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
2031  advisable to supply an \fIovector\fP.  is usually advisable to supply an \fIovector\fP of reasonable size.
2032    .P
2033    There are some cases where zero is returned (indicating vector overflow) when
2034    in fact the vector is exactly the right size for the final match. For example,
2035    consider the pattern
2036    .sp
2037      (a)(?:(b)c|bd)
2038    .sp
2039    If a vector of 6 elements (allowing for only 1 captured substring) is given
2040    with subject string "abd", \fBpcre_exec()\fP will try to set the second
2041    captured string, thereby recording a vector overflow, before failing to match
2042    "c" and backing up to try the second alternative. The zero return, however,
2043    does correctly indicate that the maximum number of slots (namely 2) have been
2044    filled. In similar cases where there is temporary overflow, but the final
2045    number of used slots is actually less than the maximum, a non-zero value is
2046    returned.
2047  .P  .P
2048  The \fBpcre_info()\fP function can be used to find out how many capturing  The \fBpcre_fullinfo()\fP function can be used to find out how many capturing
2049  subpatterns there are in a compiled pattern. The smallest size for  subpatterns there are in a compiled pattern. The smallest size for
2050  \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
2051  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.
# Line 1331  Offset values that correspond to unused Line 2061  Offset values that correspond to unused
2061  expression are also set to -1. For example, if the string "abc" is matched  expression are also set to -1. For example, if the string "abc" is matched
2062  against the pattern (abc)(x(yz)?)? subpatterns 2 and 3 are not matched. The  against the pattern (abc)(x(yz)?)? subpatterns 2 and 3 are not matched. The
2063  return from the function is 2, because the highest used capturing subpattern  return from the function is 2, because the highest used capturing subpattern
2064  number is 1. However, you can refer to the offsets for the second and third  number is 1, and the offsets for for the second and third capturing subpatterns
2065  capturing subpatterns if you wish (assuming the vector is large enough, of  (assuming the vector is large enough, of course) are set to -1.
2066  course).  .P
2067    \fBNote\fP: Elements in the first two-thirds of \fIovector\fP that do not
2068    correspond to capturing parentheses in the pattern are never changed. That is,
2069    if a pattern contains \fIn\fP capturing parentheses, no more than
2070    \fIovector[0]\fP to \fIovector[2n+1]\fP are set by \fBpcre_exec()\fP. The other
2071    elements (in the first two-thirds) retain whatever values they previously had.
2072  .P  .P
2073  Some convenience functions are provided for extracting the captured substrings  Some convenience functions are provided for extracting the captured substrings
2074  as separate strings. These are described below.  as separate strings. These are described below.
2075  .  .
2076    .
2077  .\" HTML <a name="errorlist"></a>  .\" HTML <a name="errorlist"></a>
2078  .SS "Error return values from \fBpcre_exec()\fP"  .SS "Error return values from \fBpcre_exec()\fP"
2079  .rs  .rs
# Line 1379  If a pattern contains back references, b Line 2115  If a pattern contains back references, b
2115  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
2116  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
2117  automatically freed at the end of matching.  automatically freed at the end of matching.
2118    .P
2119    This error is also given if \fBpcre_stack_malloc()\fP fails in
2120    \fBpcre_exec()\fP. This can happen only when PCRE has been compiled with
2121    \fB--disable-stack-for-recursion\fP.
2122  .sp  .sp
2123    PCRE_ERROR_NOSUBSTRING    (-7)    PCRE_ERROR_NOSUBSTRING    (-7)
2124  .sp  .sp
# Line 1403  documentation for details. Line 2143  documentation for details.
2143  .sp  .sp
2144    PCRE_ERROR_BADUTF8        (-10)    PCRE_ERROR_BADUTF8        (-10)
2145  .sp  .sp
2146  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,
2147    and the PCRE_NO_UTF8_CHECK option was not set. If the size of the output vector
2148    (\fIovecsize\fP) is at least 2, the byte offset to the start of the the invalid
2149    UTF-8 character is placed in the first element, and a reason code is placed in
2150    the second element. The reason codes are listed in the
2151    .\" HTML <a href="#badutf8reasons">
2152    .\" </a>
2153    following section.
2154    .\"
2155    For backward compatibility, if PCRE_PARTIAL_HARD is set and the problem is a
2156    truncated UTF-8 character at the end of the subject (reason codes 1 to 5),
2157    PCRE_ERROR_SHORTUTF8 is returned instead of PCRE_ERROR_BADUTF8.
2158  .sp  .sp
2159    PCRE_ERROR_BADUTF8_OFFSET (-11)    PCRE_ERROR_BADUTF8_OFFSET (-11)
2160  .sp  .sp
2161  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
2162  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
2163    \fIstartoffset\fP did not point to the beginning of a UTF-8 character or the
2164    end of the subject.
2165  .sp  .sp
2166    PCRE_ERROR_PARTIAL        (-12)    PCRE_ERROR_PARTIAL        (-12)
2167  .sp  .sp
# Line 1420  documentation for details of partial mat Line 2173  documentation for details of partial mat
2173  .sp  .sp
2174    PCRE_ERROR_BADPARTIAL     (-13)    PCRE_ERROR_BADPARTIAL     (-13)
2175  .sp  .sp
2176  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
2177  are not supported for partial matching. See the  option was used with a compiled pattern containing items that were not
2178  .\" HREF  supported for partial matching. From release 8.00 onwards, there are no
2179  \fBpcrepartial\fP  restrictions on partial matching.
 .\"  
 documentation for details of partial matching.  
2180  .sp  .sp
2181    PCRE_ERROR_INTERNAL       (-14)    PCRE_ERROR_INTERNAL       (-14)
2182  .sp  .sp
# Line 1442  The internal recursion limit, as specifi Line 2193  The internal recursion limit, as specifi
2193  field in a \fBpcre_extra\fP structure (or defaulted) was reached. See the  field in a \fBpcre_extra\fP structure (or defaulted) was reached. See the
2194  description above.  description above.
2195  .sp  .sp
   PCRE_ERROR_NULLWSLIMIT    (-22)  
 .sp  
 When a group that can match an empty substring is repeated with an unbounded  
 upper limit, the subject position at the start of the group must be remembered,  
 so that a test for an empty string can be made when the end of the group is  
 reached. Some workspace is required for this; if it runs out, this error is  
 given.  
 .sp  
2196    PCRE_ERROR_BADNEWLINE     (-23)    PCRE_ERROR_BADNEWLINE     (-23)
2197  .sp  .sp
2198  An invalid combination of PCRE_NEWLINE_\fIxxx\fP options was given.  An invalid combination of PCRE_NEWLINE_\fIxxx\fP options was given.
2199    .sp
2200      PCRE_ERROR_BADOFFSET      (-24)
2201    .sp
2202    The value of \fIstartoffset\fP was negative or greater than the length of the
2203    subject, that is, the value in \fIlength\fP.
2204    .sp
2205      PCRE_ERROR_SHORTUTF8      (-25)
2206    .sp
2207    This error is returned instead of PCRE_ERROR_BADUTF8 when the subject string
2208    ends with a truncated UTF-8 character and the PCRE_PARTIAL_HARD option is set.
2209    Information about the failure is returned as for PCRE_ERROR_BADUTF8. It is in
2210    fact sufficient to detect this case, but this special error code for
2211    PCRE_PARTIAL_HARD precedes the implementation of returned information; it is
2212    retained for backwards compatibility.
2213    .sp
2214      PCRE_ERROR_RECURSELOOP    (-26)
2215    .sp
2216    This error is returned when \fBpcre_exec()\fP detects a recursion loop within
2217    the pattern. Specifically, it means that either the whole pattern or a
2218    subpattern has been called recursively for the second time at the same position
2219    in the subject string. Some simple patterns that might do this are detected and
2220    faulted at compile time, but more complicated cases, in particular mutual
2221    recursions between two different subpatterns, cannot be detected until run
2222    time.
2223    .sp
2224      PCRE_ERROR_JIT_STACKLIMIT (-27)
2225    .sp
2226    This error is returned when a pattern that was successfully studied using a
2227    JIT compile option is being matched, but the memory available for the
2228    just-in-time processing stack is not large enough. See the
2229    .\" HREF
2230    \fBpcrejit\fP
2231    .\"
2232    documentation for more details.
2233    .sp
2234      PCRE_ERROR_BADMODE        (-28)
2235    .sp
2236    This error is given if a pattern that was compiled by the 8-bit library is
2237    passed to a 16-bit or 32-bit library function, or vice versa.
2238    .sp
2239      PCRE_ERROR_BADENDIANNESS  (-29)
2240    .sp
2241    This error is given if a pattern that was compiled and saved is reloaded on a
2242    host with different endianness. The utility function
2243    \fBpcre_pattern_to_host_byte_order()\fP can be used to convert such a pattern
2244    so that it runs on the new host.
2245    .sp
2246      PCRE_ERROR_JIT_BADOPTION
2247    .sp
2248    This error is returned when a pattern that was successfully studied using a JIT
2249    compile option is being matched, but the matching mode (partial or complete
2250    match) does not correspond to any JIT compilation mode. When the JIT fast path
2251    function is used, this error may be also given for invalid options. See the
2252    .\" HREF
2253    \fBpcrejit\fP
2254    .\"
2255    documentation for more details.
2256    .sp
2257      PCRE_ERROR_BADLENGTH      (-32)
2258    .sp
2259    This error is given if \fBpcre_exec()\fP is called with a negative value for
2260    the \fIlength\fP argument.
2261  .P  .P
2262  Error numbers -16 to -20 are not used by \fBpcre_exec()\fP.  Error numbers -16 to -20, -22, and 30 are not used by \fBpcre_exec()\fP.
2263    .
2264    .
2265    .\" HTML <a name="badutf8reasons"></a>
2266    .SS "Reason codes for invalid UTF-8 strings"
2267    .rs
2268    .sp
2269    This section applies only to the 8-bit library. The corresponding information
2270    for the 16-bit library is given in the
2271    .\" HREF
2272    \fBpcre16\fP
2273    .\"
2274    page. The corresponding information for the 32-bit library is given in the
2275    .\" HREF
2276    \fBpcre32\fP
2277    .\"
2278    page.
2279    .P
2280    When \fBpcre_exec()\fP returns either PCRE_ERROR_BADUTF8 or
2281    PCRE_ERROR_SHORTUTF8, and the size of the output vector (\fIovecsize\fP) is at
2282    least 2, the offset of the start of the invalid UTF-8 character is placed in
2283    the first output vector element (\fIovector[0]\fP) and a reason code is placed
2284    in the second element (\fIovector[1]\fP). The reason codes are given names in
2285    the \fBpcre.h\fP header file:
2286    .sp
2287      PCRE_UTF8_ERR1
2288      PCRE_UTF8_ERR2
2289      PCRE_UTF8_ERR3
2290      PCRE_UTF8_ERR4
2291      PCRE_UTF8_ERR5
2292    .sp
2293    The string ends with a truncated UTF-8 character; the code specifies how many
2294    bytes are missing (1 to 5). Although RFC 3629 restricts UTF-8 characters to be
2295    no longer than 4 bytes, the encoding scheme (originally defined by RFC 2279)
2296    allows for up to 6 bytes, and this is checked first; hence the possibility of
2297    4 or 5 missing bytes.
2298    .sp
2299      PCRE_UTF8_ERR6
2300      PCRE_UTF8_ERR7
2301      PCRE_UTF8_ERR8
2302      PCRE_UTF8_ERR9
2303      PCRE_UTF8_ERR10
2304    .sp
2305    The two most significant bits of the 2nd, 3rd, 4th, 5th, or 6th byte of the
2306    character do not have the binary value 0b10 (that is, either the most
2307    significant bit is 0, or the next bit is 1).
2308    .sp
2309      PCRE_UTF8_ERR11
2310      PCRE_UTF8_ERR12
2311    .sp
2312    A character that is valid by the RFC 2279 rules is either 5 or 6 bytes long;
2313    these code points are excluded by RFC 3629.
2314    .sp
2315      PCRE_UTF8_ERR13
2316    .sp
2317    A 4-byte character has a value greater than 0x10fff; these code points are
2318    excluded by RFC 3629.
2319    .sp
2320      PCRE_UTF8_ERR14
2321    .sp
2322    A 3-byte character has a value in the range 0xd800 to 0xdfff; this range of
2323    code points are reserved by RFC 3629 for use with UTF-16, and so are excluded
2324    from UTF-8.
2325    .sp
2326      PCRE_UTF8_ERR15
2327      PCRE_UTF8_ERR16
2328      PCRE_UTF8_ERR17
2329      PCRE_UTF8_ERR18
2330      PCRE_UTF8_ERR19
2331    .sp
2332    A 2-, 3-, 4-, 5-, or 6-byte character is "overlong", that is, it codes for a
2333    value that can be represented by fewer bytes, which is invalid. For example,
2334    the two bytes 0xc0, 0xae give the value 0x2e, whose correct coding uses just
2335    one byte.
2336    .sp
2337      PCRE_UTF8_ERR20
2338    .sp
2339    The two most significant bits of the first byte of a character have the binary
2340    value 0b10 (that is, the most significant bit is 1 and the second is 0). Such a
2341    byte can only validly occur as the second or subsequent byte of a multi-byte
2342    character.
2343    .sp
2344      PCRE_UTF8_ERR21
2345    .sp
2346    The first byte of a character has the value 0xfe or 0xff. These values can
2347    never occur in a valid UTF-8 string.
2348    .sp
2349      PCRE_UTF8_ERR2
2350    .sp
2351    Non-character. These are the last two characters in each plane (0xfffe, 0xffff,
2352    0x1fffe, 0x1ffff .. 0x10fffe, 0x10ffff), and the characters 0xfdd0..0xfdef.
2353  .  .
2354  .  .
2355  .SH "EXTRACTING CAPTURED SUBSTRINGS BY NUMBER"  .SH "EXTRACTING CAPTURED SUBSTRINGS BY NUMBER"
# Line 1603  These functions call \fBpcre_get_stringn Line 2498  These functions call \fBpcre_get_stringn
2498  then call \fBpcre_copy_substring()\fP or \fBpcre_get_substring()\fP, as  then call \fBpcre_copy_substring()\fP or \fBpcre_get_substring()\fP, as
2499  appropriate. \fBNOTE:\fP If PCRE_DUPNAMES is set and there are duplicate names,  appropriate. \fBNOTE:\fP If PCRE_DUPNAMES is set and there are duplicate names,
2500  the behaviour may not be what you want (see the next section).  the behaviour may not be what you want (see the next section).
2501    .P
2502    \fBWarning:\fP If the pattern uses the (?| feature to set up multiple
2503    subpatterns with the same number, as described in the
2504    .\" HTML <a href="pcrepattern.html#dupsubpatternnumber">
2505    .\" </a>
2506    section on duplicate subpattern numbers
2507    .\"
2508    in the
2509    .\" HREF
2510    \fBpcrepattern\fP
2511    .\"
2512    page, you cannot use names to distinguish the different subpatterns, because
2513    names are not included in the compiled code. The matching process uses only
2514    numbers. For this reason, the use of different names for subpatterns of the
2515    same number causes an error at compile time.
2516  .  .
2517  .  .
2518  .SH "DUPLICATE SUBPATTERN NAMES"  .SH "DUPLICATE SUBPATTERN NAMES"
# Line 1613  the behaviour may not be what you want ( Line 2523  the behaviour may not be what you want (
2523  .B const char *\fIname\fP, char **\fIfirst\fP, char **\fIlast\fP);  .B const char *\fIname\fP, char **\fIfirst\fP, char **\fIlast\fP);
2524  .PP  .PP
2525  When a pattern is compiled with the PCRE_DUPNAMES option, names for subpatterns  When a pattern is compiled with the PCRE_DUPNAMES option, names for subpatterns
2526  are not required to be unique. Normally, patterns with duplicate names are such  are not required to be unique. (Duplicate names are always allowed for
2527  that in any one match, only one of the named subpatterns participates. An  subpatterns with the same number, created by using the (?| feature. Indeed, if
2528  example is shown in the  such subpatterns are named, they are required to use the same names.)
2529    .P
2530    Normally, patterns with duplicate names are such that in any one match, only
2531    one of the named subpatterns participates. An example is shown in the
2532  .\" HREF  .\" HREF
2533  \fBpcrepattern\fP  \fBpcrepattern\fP
2534  .\"  .\"
2535  documentation. When duplicates are present, \fBpcre_copy_named_substring()\fP  documentation.
2536  and \fBpcre_get_named_substring()\fP return the first substring corresponding  .P
2537  to the given name that is set. If none are set, an empty string is returned.  When duplicates are present, \fBpcre_copy_named_substring()\fP and
2538  The \fBpcre_get_stringnumber()\fP function returns one of the numbers that are  \fBpcre_get_named_substring()\fP return the first substring corresponding to
2539  associated with the name, but it is not defined which it is.  the given name that is set. If none are set, PCRE_ERROR_NOSUBSTRING (-7) is
2540  .sp  returned; no data is returned. The \fBpcre_get_stringnumber()\fP function
2541    returns one of the numbers that are associated with the name, but it is not
2542    defined which it is.
2543    .P
2544  If you want to get full details of all captured substrings for a given name,  If you want to get full details of all captured substrings for a given name,
2545  you must use the \fBpcre_get_stringtable_entries()\fP function. The first  you must use the \fBpcre_get_stringtable_entries()\fP function. The first
2546  argument is the compiled pattern, and the second is the name. The third and  argument is the compiled pattern, and the second is the name. The third and
# Line 1632  fourth are pointers to variables which a Line 2548  fourth are pointers to variables which a
2548  has run, they point to the first and last entries in the name-to-number table  has run, they point to the first and last entries in the name-to-number table
2549  for the given name. The function itself returns the length of each entry, or  for the given name. The function itself returns the length of each entry, or
2550  PCRE_ERROR_NOSUBSTRING (-7) if there are none. The format of the table is  PCRE_ERROR_NOSUBSTRING (-7) if there are none. The format of the table is
2551  described above in the section entitled \fIInformation about a pattern\fP.  described above in the section entitled \fIInformation about a pattern\fP
2552    .\" HTML <a href="#infoaboutpattern">
2553    .\" </a>
2554    above.
2555    .\"
2556  Given all the relevant entries for the name, you can extract each of their  Given all the relevant entries for the name, you can extract each of their
2557  numbers, and hence the captured data, if any.  numbers, and hence the captured data, if any.
2558  .  .
# Line 1659  other alternatives. Ultimately, when it Line 2579  other alternatives. Ultimately, when it
2579  will yield PCRE_ERROR_NOMATCH.  will yield PCRE_ERROR_NOMATCH.
2580  .  .
2581  .  .
2582    .SH "OBTAINING AN ESTIMATE OF STACK USAGE"
2583    .rs
2584    .sp
2585    Matching certain patterns using \fBpcre_exec()\fP can use a lot of process
2586    stack, which in certain environments can be rather limited in size. Some users
2587    find it helpful to have an estimate of the amount of stack that is used by
2588    \fBpcre_exec()\fP, to help them set recursion limits, as described in the
2589    .\" HREF
2590    \fBpcrestack\fP
2591    .\"
2592    documentation. The estimate that is output by \fBpcretest\fP when called with
2593    the \fB-m\fP and \fB-C\fP options is obtained by calling \fBpcre_exec\fP with
2594    the values NULL, NULL, NULL, -999, and -999 for its first five arguments.
2595    .P
2596    Normally, if its first argument is NULL, \fBpcre_exec()\fP immediately returns
2597    the negative error code PCRE_ERROR_NULL, but with this special combination of
2598    arguments, it returns instead a negative number whose absolute value is the
2599    approximate stack frame size in bytes. (A negative number is used so that it is
2600    clear that no match has happened.) The value is approximate because in some
2601    cases, recursive calls to \fBpcre_exec()\fP occur when there are one or two
2602    additional variables on the stack.
2603    .P
2604    If PCRE has been compiled to use the heap instead of the stack for recursion,
2605    the value returned is the size of each block that is obtained from the heap.
2606    .
2607    .
2608  .\" HTML <a name="dfamatch"></a>  .\" HTML <a name="dfamatch"></a>
2609  .SH "MATCHING A PATTERN: THE ALTERNATIVE FUNCTION"  .SH "MATCHING A PATTERN: THE ALTERNATIVE FUNCTION"
2610  .rs  .rs
# Line 1676  a compiled pattern, using a matching alg Line 2622  a compiled pattern, using a matching alg
2622  just once, and does not backtrack. This has different characteristics to the  just once, and does not backtrack. This has different characteristics to the
2623  normal algorithm, and is not compatible with Perl. Some of the features of PCRE  normal algorithm, and is not compatible with Perl. Some of the features of PCRE
2624  patterns are not supported. Nevertheless, there are times when this kind of  patterns are not supported. Nevertheless, there are times when this kind of
2625  matching can be useful. For a discussion of the two matching algorithms, see  matching can be useful. For a discussion of the two matching algorithms, and a
2626  the  list of features that \fBpcre_dfa_exec()\fP does not support, see the
2627  .\" HREF  .\" HREF
2628  \fBpcrematching\fP  \fBpcrematching\fP
2629  .\"  .\"
# Line 1716  Here is an example of a simple call to \ Line 2662  Here is an example of a simple call to \
2662  .sp  .sp
2663  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
2664  zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_\fIxxx\fP,  zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_\fIxxx\fP,
2665  PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NO_UTF8_CHECK, PCRE_PARTIAL,  PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NOTEMPTY_ATSTART,
2666  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,
2667  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.
2668  .sp  All but the last four of these are exactly the same as for \fBpcre_exec()\fP,
2669    PCRE_PARTIAL  so their description is not repeated here.
2670  .sp  .sp
2671  This has the same general effect as it does for \fBpcre_exec()\fP, but the    PCRE_PARTIAL_HARD
2672  details are slightly different. When PCRE_PARTIAL is set for    PCRE_PARTIAL_SOFT
2673  \fBpcre_dfa_exec()\fP, the return code PCRE_ERROR_NOMATCH is converted into  .sp
2674  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
2675  complete matches, but there is still at least one matching possibility. The  details are slightly different. When PCRE_PARTIAL_HARD is set for
2676  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
2677  matching string.  is reached and there is still at least one matching possibility that requires
2678    additional characters. This happens even if some complete matches have also
2679    been found. When PCRE_PARTIAL_SOFT is set, the return code PCRE_ERROR_NOMATCH
2680    is converted into PCRE_ERROR_PARTIAL if the end of the subject is reached,
2681    there have been no complete matches, but there is still at least one matching
2682    possibility. The portion of the string that was inspected when the longest
2683    partial match was found is set as the first matching string in both cases.
2684    There is a more detailed discussion of partial and multi-segment matching, with
2685    examples, in the
2686    .\" HREF
2687    \fBpcrepartial\fP
2688    .\"
2689    documentation.
2690  .sp  .sp
2691    PCRE_DFA_SHORTEST    PCRE_DFA_SHORTEST
2692  .sp  .sp
# Line 1739  matching point in the subject string. Line 2697  matching point in the subject string.
2697  .sp  .sp
2698    PCRE_DFA_RESTART    PCRE_DFA_RESTART
2699  .sp  .sp
2700  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
2701  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
2702  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
2703  option requests this action; when it is set, the \fIworkspace\fP and  \fIworkspace\fP and \fIwscount\fP options must reference the same vector as
2704  \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
2705  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  
2706  .\" HREF  .\" HREF
2707  \fBpcrepartial\fP  \fBpcrepartial\fP
2708  .\"  .\"
2709  documentation.  documentation.
2710  .  .
2711    .
2712  .SS "Successful returns from \fBpcre_dfa_exec()\fP"  .SS "Successful returns from \fBpcre_dfa_exec()\fP"
2713  .rs  .rs
2714  .sp  .sp
# Line 1782  returns data, even though the meaning of Line 2740  returns data, even though the meaning of
2740  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
2741  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
2742  \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
2743  the longest matches.  the longest matches. Unlike \fBpcre_exec()\fP, \fBpcre_dfa_exec()\fP can use
2744    the entire \fIovector\fP for returning matched strings.
2745    .
2746  .  .
2747  .SS "Error returns from \fBpcre_dfa_exec()\fP"  .SS "Error returns from \fBpcre_dfa_exec()\fP"
2748  .rs  .rs
# Line 1811  group. These are not supported. Line 2771  group. These are not supported.
2771    PCRE_ERROR_DFA_UMLIMIT    (-18)    PCRE_ERROR_DFA_UMLIMIT    (-18)
2772  .sp  .sp
2773  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
2774  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
2775  supported (it is meaningless).  \fImatch_limit_recursion\fP fields. This is not supported (these fields are
2776    meaningless for DFA matching).
2777  .sp  .sp
2778    PCRE_ERROR_DFA_WSSIZE     (-19)    PCRE_ERROR_DFA_WSSIZE     (-19)
2779  .sp  .sp
# Line 1825  When a recursive subpattern is processed Line 2786  When a recursive subpattern is processed
2786  recursively, using private vectors for \fIovector\fP and \fIworkspace\fP. This  recursively, using private vectors for \fIovector\fP and \fIworkspace\fP. This
2787  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
2788  extremely rare, as a vector of size 1000 is used.  extremely rare, as a vector of size 1000 is used.
2789    .sp
2790      PCRE_ERROR_DFA_BADRESTART (-30)
2791    .sp
2792    When \fBpcre_dfa_exec()\fP is called with the \fBPCRE_DFA_RESTART\fP option,
2793    some plausibility checks are made on the contents of the workspace, which
2794    should contain data about the previous partial match. If any of these checks
2795    fail, this error is given.
2796  .  .
2797  .  .
2798  .SH "SEE ALSO"  .SH "SEE ALSO"
2799  .rs  .rs
2800  .sp  .sp
2801  \fBpcrebuild\fP(3), \fBpcrecallout\fP(3), \fBpcrecpp(3)\fP(3),  \fBpcre16\fP(3), \fBpcre32\fP(3), \fBpcrebuild\fP(3), \fBpcrecallout\fP(3),
2802  \fBpcrematching\fP(3), \fBpcrepartial\fP(3), \fBpcreposix\fP(3),  \fBpcrecpp(3)\fP(3), \fBpcrematching\fP(3), \fBpcrepartial\fP(3),
2803  \fBpcreprecompile\fP(3), \fBpcresample\fP(3), \fBpcrestack\fP(3).  \fBpcreposix\fP(3), \fBpcreprecompile\fP(3), \fBpcresample\fP(3),
2804    \fBpcrestack\fP(3).
2805  .  .
2806  .  .
2807  .SH AUTHOR  .SH AUTHOR
# Line 1849  Cambridge CB2 3QH, England. Line 2818  Cambridge CB2 3QH, England.
2818  .rs  .rs
2819  .sp  .sp
2820  .nf  .nf
2821  Last updated: 16 April 2007  Last updated: 31 October 2012
2822  Copyright (c) 1997-2007 University of Cambridge.  Copyright (c) 1997-2012 University of Cambridge.
2823  .fi  .fi

Legend:
Removed from v.150  
changed lines
  Added in v.1194

  ViewVC Help
Powered by ViewVC 1.1.5