/[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 442 by ph10, Fri Sep 11 10:21:02 2009 UTC revision 1080 by chpe, Tue Oct 16 15:55:07 2012 UTC
# Line 1  Line 1 
1  .TH PCREAPI 3  .TH PCREAPI 3 "07 September 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 pcre_jit_stack *pcre_jit_stack_alloc(int \fIstartsize\fP, int \fImaxsize\fP);
99    .PP
100    .B void pcre_jit_stack_free(pcre_jit_stack *\fIstack\fP);
101    .PP
102    .B void pcre_assign_jit_stack(pcre_extra *\fIextra\fP,
103    .ti +5n
104    .B pcre_jit_callback \fIcallback\fP, void *\fIdata\fP);
105  .PP  .PP
106  .B const unsigned char *pcre_maketables(void);  .B const unsigned char *pcre_maketables(void);
107  .PP  .PP
# Line 89  PCRE - Perl-compatible regular expressio Line 109  PCRE - Perl-compatible regular expressio
109  .ti +5n  .ti +5n
110  .B int \fIwhat\fP, void *\fIwhere\fP);  .B int \fIwhat\fP, void *\fIwhere\fP);
111  .PP  .PP
 .B int pcre_info(const pcre *\fIcode\fP, int *\fIoptptr\fP, int  
 .B *\fIfirstcharptr\fP);  
 .PP  
112  .B int pcre_refcount(pcre *\fIcode\fP, int \fIadjust\fP);  .B int pcre_refcount(pcre *\fIcode\fP, int \fIadjust\fP);
113  .PP  .PP
114  .B int pcre_config(int \fIwhat\fP, void *\fIwhere\fP);  .B int pcre_config(int \fIwhat\fP, void *\fIwhere\fP);
115  .PP  .PP
116  .B char *pcre_version(void);  .B const char *pcre_version(void);
117  .PP  .PP
118    .B int pcre_pattern_to_host_byte_order(pcre *\fIcode\fP,
119    .ti +5n
120    .B pcre_extra *\fIextra\fP, const unsigned char *\fItables\fP);
121    .
122    .
123    .SH "PCRE NATIVE API INDIRECTED FUNCTIONS"
124    .rs
125    .sp
126  .B void *(*pcre_malloc)(size_t);  .B void *(*pcre_malloc)(size_t);
127  .PP  .PP
128  .B void (*pcre_free)(void *);  .B void (*pcre_free)(void *);
# Line 109  PCRE - Perl-compatible regular expressio Line 134  PCRE - Perl-compatible regular expressio
134  .B int (*pcre_callout)(pcre_callout_block *);  .B int (*pcre_callout)(pcre_callout_block *);
135  .  .
136  .  .
137    .SH "PCRE 8-BIT, 16-BIT AND 32-BIT LIBRARIES"
138    .rs
139    .sp
140    From release 8.30, PCRE can be compiled as a library for handling 16-bit
141    character strings as well as, or instead of, the original library that handles
142    8-bit character strings. From release 8.FIXME, PCRE can also be compiled as a
143    library for handling 32-bit character strings. To avoid too much complication,
144    this document describes the 8-bit versions of the functions, with only
145    occasional references to the 16-bit and 32-bit libraries.
146    .P
147    The 16-bit functions operate in the same way as their 8-bit counterparts; they
148    just use different data types for their arguments and results, and their names
149    start with \fBpcre16_\fP instead of \fBpcre_\fP. For every option that has UTF8
150    in its name (for example, PCRE_UTF8), there is a corresponding 16-bit name with
151    UTF8 replaced by UTF16. This facility is in fact just cosmetic; the 16-bit
152    option names define the same bit values.
153    .P
154    The 32-bit functions operate in the same way as their 8-bit counterparts; they
155    just use different data types for their arguments and results, and their names
156    start with \fBpcre32_\fP instead of \fBpcre_\fP. For every option that has UTF8
157    in its name (for example, PCRE_UTF8), there is a corresponding 32-bit name with
158    UTF8 replaced by UTF32. This facility is in fact just cosmetic; the 32-bit
159    option names define the same bit values.
160    .P
161    References to bytes and UTF-8 in this document should be read as references to
162    16-bit data quantities and UTF-16 when using the 16-bit library, unless
163    specified otherwise. More details of the specific differences for the 16-bit
164    library are given in the
165    .\" HREF
166    \fBpcre16\fP
167    .\"
168    page.
169    .
170    .P
171    References to bytes and UTF-8 in this document should be read as references to
172    32-bit data quantities and UTF-32 when using the 32-bit library, unless
173    specified otherwise. More details of the specific differences for the 32-bit
174    library are given in the
175    .\" HREF
176    \fBpcre32\fP
177    .\"
178    page.
179    .
180    .
181  .SH "PCRE API OVERVIEW"  .SH "PCRE API OVERVIEW"
182  .rs  .rs
183  .sp  .sp
184  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
185  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
186  API. These are described in the  POSIX regular expression API, but they do not give access to all the
187    functionality. They are described in the
188  .\" HREF  .\" HREF
189  \fBpcreposix\fP  \fBpcreposix\fP
190  .\"  .\"
191  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++
192  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
193    documented in the
194  .\" HREF  .\" HREF
195  \fBpcrecpp\fP  \fBpcrecpp\fP
196  .\"  .\"
197  page.  page.
198  .P  .P
199  The native API C function prototypes are defined in the header file  The native API C function prototypes are defined in the header file
200  \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
201  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
202  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
203  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
204  Applications can use these to include support for different releases of PCRE.  for the library. Applications can use these to include support for different
205    releases of PCRE.
206    .P
207    In a Windows environment, if you want to statically link an application program
208    against a non-dll \fBpcre.a\fP file, you must define PCRE_STATIC before
209    including \fBpcre.h\fP or \fBpcrecpp.h\fP, because otherwise the
210    \fBpcre_malloc()\fP and \fBpcre_free()\fP exported functions will be declared
211    \fB__declspec(dllimport)\fP, with unwanted results.
212  .P  .P
213  The functions \fBpcre_compile()\fP, \fBpcre_compile2()\fP, \fBpcre_study()\fP,  The functions \fBpcre_compile()\fP, \fBpcre_compile2()\fP, \fBpcre_study()\fP,
214  and \fBpcre_exec()\fP are used for compiling and matching regular expressions  and \fBpcre_exec()\fP are used for compiling and matching regular expressions
# Line 146  documentation, and the Line 224  documentation, and the
224  .\"  .\"
225  documentation describes how to compile and run it.  documentation describes how to compile and run it.
226  .P  .P
227    Just-in-time compiler support is an optional feature of PCRE that can be built
228    in appropriate hardware environments. It greatly speeds up the matching
229    performance of many patterns. Simple programs can easily request that it be
230    used if available, by setting an option that is ignored when it is not
231    relevant. More complicated programs might need to make use of the functions
232    \fBpcre_jit_stack_alloc()\fP, \fBpcre_jit_stack_free()\fP, and
233    \fBpcre_assign_jit_stack()\fP in order to control the JIT code's memory usage.
234    These functions are discussed in the
235    .\" HREF
236    \fBpcrejit\fP
237    .\"
238    documentation.
239    .P
240  A second matching function, \fBpcre_dfa_exec()\fP, which is not  A second matching function, \fBpcre_dfa_exec()\fP, which is not
241  Perl-compatible, is also provided. This uses a different algorithm for the  Perl-compatible, is also provided. This uses a different algorithm for the
242  matching. The alternative algorithm finds all possible matches (at a given  matching. The alternative algorithm finds all possible matches (at a given
# Line 180  specialist use. Most commonly, no specia Line 271  specialist use. Most commonly, no specia
271  internal tables that are generated when PCRE is built are used.  internal tables that are generated when PCRE is built are used.
272  .P  .P
273  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
274  compiled pattern; \fBpcre_info()\fP is an obsolete version that returns only  compiled pattern. The function \fBpcre_version()\fP returns a pointer to a
275  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.  
276  .P  .P
277  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
278  containing a compiled pattern. This is provided for the benefit of  containing a compiled pattern. This is provided for the benefit of
# Line 231  PCRE supports five different conventions Line 320  PCRE supports five different conventions
320  strings: a single CR (carriage return) character, a single LF (linefeed)  strings: a single CR (carriage return) character, a single LF (linefeed)
321  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
322  Unicode newline sequence. The Unicode newline sequences are the three just  Unicode newline sequence. The Unicode newline sequences are the three just
323  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,
324  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
325  (paragraph separator, U+2029).  (paragraph separator, U+2029).
326  .P  .P
# Line 276  callout function pointed to by \fBpcre_c Line 365  callout function pointed to by \fBpcre_c
365  .P  .P
366  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
367  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.
368    .P
369    If the just-in-time optimization feature is being used, it needs separate
370    memory stack areas for each thread. See the
371    .\" HREF
372    \fBpcrejit\fP
373    .\"
374    documentation for more details.
375  .  .
376  .  .
377  .SH "SAVING PRECOMPILED PATTERNS FOR LATER USE"  .SH "SAVING PRECOMPILED PATTERNS FOR LATER USE"
# Line 287  which it was compiled. Details are given Line 383  which it was compiled. Details are given
383  .\" HREF  .\" HREF
384  \fBpcreprecompile\fP  \fBpcreprecompile\fP
385  .\"  .\"
386  documentation. However, compiling a regular expression with one version of PCRE  documentation, which includes a description of the
387  for use with a different version is not guaranteed to work and may cause  \fBpcre_pattern_to_host_byte_order()\fP function. However, compiling a regular
388  crashes.  expression with one version of PCRE for use with a different version is not
389    guaranteed to work and may cause crashes.
390  .  .
391  .  .
392  .SH "CHECKING BUILD-TIME OPTIONS"  .SH "CHECKING BUILD-TIME OPTIONS"
# Line 306  documentation has more details about the Line 403  documentation has more details about the
403  .P  .P
404  The first argument for \fBpcre_config()\fP is an integer, specifying which  The first argument for \fBpcre_config()\fP is an integer, specifying which
405  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
406  which the information is placed. The following information is available:  which the information is placed. The returned value is zero on success, or the
407    negative error code PCRE_ERROR_BADOPTION if the value in the first argument is
408    not recognized. The following information is available:
409  .sp  .sp
410    PCRE_CONFIG_UTF8    PCRE_CONFIG_UTF8
411  .sp  .sp
412  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;
413  otherwise it is set to zero.  otherwise it is set to zero. This value should normally be given to the 8-bit
414    version of this function, \fBpcre_config()\fP. If it is given to the 16-bit
415    or 32-bit version of this function, the result is PCRE_ERROR_BADOPTION.
416    .sp
417      PCRE_CONFIG_UTF16
418    .sp
419    The output is an integer that is set to one if UTF-16 support is available;
420    otherwise it is set to zero. This value should normally be given to the 16-bit
421    version of this function, \fBpcre16_config()\fP. If it is given to the 8-bit
422    or 32-bit version of this function, the result is PCRE_ERROR_BADOPTION.
423    .sp
424      PCRE_CONFIG_UTF32
425    .sp
426    The output is an integer that is set to one if UTF-32 support is available;
427    otherwise it is set to zero. This value should normally be given to the 32-bit
428    version of this function, \fBpcre32_config()\fP. If it is given to the 8-bit
429    or 16-bit version of this function, the result is PCRE_ERROR_BADOPTION.
430  .sp  .sp
431    PCRE_CONFIG_UNICODE_PROPERTIES    PCRE_CONFIG_UNICODE_PROPERTIES
432  .sp  .sp
433  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
434  properties is available; otherwise it is set to zero.  properties is available; otherwise it is set to zero.
435  .sp  .sp
436      PCRE_CONFIG_JIT
437    .sp
438    The output is an integer that is set to one if support for just-in-time
439    compiling is available; otherwise it is set to zero.
440    .sp
441      PCRE_CONFIG_JITTARGET
442    .sp
443    The output is a pointer to a zero-terminated "const char *" string. If JIT
444    support is available, the string contains the name of the architecture for
445    which the JIT compiler is configured, for example "x86 32bit (little endian +
446    unaligned)". If JIT support is not available, the result is NULL.
447    .sp
448    PCRE_CONFIG_NEWLINE    PCRE_CONFIG_NEWLINE
449  .sp  .sp
450  The output is an integer whose value specifies the default character sequence  The output is an integer whose value specifies the default character sequence
451  that is recognized as meaning "newline". The four values that are supported  that is recognized as meaning "newline". The values that are supported in
452  are: 10 for LF, 13 for CR, 3338 for CRLF, -2 for ANYCRLF, and -1 for ANY.  ASCII/Unicode environments are: 10 for LF, 13 for CR, 3338 for CRLF, -2 for
453  Though they are derived from ASCII, the same values are returned in EBCDIC  ANYCRLF, and -1 for ANY. In EBCDIC environments, CR, ANYCRLF, and ANY yield the
454  environments. The default should normally correspond to the standard sequence  same values. However, the value for LF is normally 21, though some EBCDIC
455  for your operating system.  environments use 37. The corresponding values for CRLF are 3349 and 3365. The
456    default should normally correspond to the standard sequence for your operating
457    system.
458  .sp  .sp
459    PCRE_CONFIG_BSR    PCRE_CONFIG_BSR
460  .sp  .sp
# Line 337  or CRLF. The default can be overridden w Line 466  or CRLF. The default can be overridden w
466    PCRE_CONFIG_LINK_SIZE    PCRE_CONFIG_LINK_SIZE
467  .sp  .sp
468  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
469  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
470  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
471  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
472  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
473    most massive patterns, since it allows the compiled pattern to be up to 64K in
474    size. Larger values allow larger regular expressions to be compiled, at the
475    expense of slower matching.
476  .sp  .sp
477    PCRE_CONFIG_POSIX_MALLOC_THRESHOLD    PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
478  .sp  .sp
# Line 395  avoiding the use of the stack. Line 527  avoiding the use of the stack.
527  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
528  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
529  the two interfaces is that \fBpcre_compile2()\fP has an additional argument,  the two interfaces is that \fBpcre_compile2()\fP has an additional argument,
530  \fIerrorcodeptr\fP, via which a numerical error code can be returned.  \fIerrorcodeptr\fP, via which a numerical error code can be returned. To avoid
531    too much repetition, we refer just to \fBpcre_compile()\fP below, but the
532    information applies equally to \fBpcre_compile2()\fP.
533  .P  .P
534  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
535  \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 412  argument, which is an address (see below Line 546  argument, which is an address (see below
546  The \fIoptions\fP argument contains various bit settings that affect the  The \fIoptions\fP argument contains various bit settings that affect the
547  compilation. It should be zero if no options are required. The available  compilation. It should be zero if no options are required. The available
548  options are described below. Some of them (in particular, those that are  options are described below. Some of them (in particular, those that are
549  compatible with Perl, but also some others) can also be set and unset from  compatible with Perl, but some others as well) can also be set and unset from
550  within the pattern (see the detailed description in the  within the pattern (see the detailed description in the
551  .\" HREF  .\" HREF
552  \fBpcrepattern\fP  \fBpcrepattern\fP
553  .\"  .\"
554  documentation). For those options that can be different in different parts of  documentation). For those options that can be different in different parts of
555  the pattern, the contents of the \fIoptions\fP argument specifies their initial  the pattern, the contents of the \fIoptions\fP argument specifies their
556  settings at the start of compilation and execution. The PCRE_ANCHORED and  settings at the start of compilation and execution. The PCRE_ANCHORED,
557  PCRE_NEWLINE_\fIxxx\fP options can be set at the time of matching as well as at  PCRE_BSR_\fIxxx\fP, PCRE_NEWLINE_\fIxxx\fP, PCRE_NO_UTF8_CHECK, and
558    PCRE_NO_START_OPTIMIZE options can be set at the time of matching as well as at
559  compile time.  compile time.
560  .P  .P
561  If \fIerrptr\fP is NULL, \fBpcre_compile()\fP returns NULL immediately.  If \fIerrptr\fP is NULL, \fBpcre_compile()\fP returns NULL immediately.
562  Otherwise, if compilation of a pattern fails, \fBpcre_compile()\fP returns  Otherwise, if compilation of a pattern fails, \fBpcre_compile()\fP returns
563  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
564  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
565  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
566  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
567  \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
568    immediate error is given). However, for an invalid UTF-8 string, the offset is
569    that of the first byte of the failing character.
570    .P
571    Some errors are not detected until the whole pattern has been scanned; in these
572    cases, the offset passed back is the length of the pattern. Note that the
573    offset is in bytes, not characters, even in UTF-8 mode. It may sometimes point
574    into the middle of a UTF-8 character.
575  .P  .P
576  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
577  \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 508  pattern. Line 650  pattern.
650  .sp  .sp
651    PCRE_DOTALL    PCRE_DOTALL
652  .sp  .sp
653  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
654  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
655  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,
656  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
657  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
658  the setting of this option.  (?s) option setting. A negative class such as [^a] always matches newline
659    characters, independent of the setting of this option.
660  .sp  .sp
661    PCRE_DUPNAMES    PCRE_DUPNAMES
662  .sp  .sp
# Line 528  documentation. Line 671  documentation.
671  .sp  .sp
672    PCRE_EXTENDED    PCRE_EXTENDED
673  .sp  .sp
674  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
675  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
676  include the VT character (code 11). In addition, characters between an  include the VT character (code 11). In addition, characters between an
677  unescaped # outside a character class and the next newline, inclusive, are also  unescaped # outside a character class and the next newline, inclusive, are also
678  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
679  pattern by a (?x) option setting.  pattern by a (?x) option setting.
680  .P  .P
681    Which characters are interpreted as newlines is controlled by the options
682    passed to \fBpcre_compile()\fP or by a special sequence at the start of the
683    pattern, as described in the section entitled
684    .\" HTML <a href="pcrepattern.html#newlines">
685    .\" </a>
686    "Newline conventions"
687    .\"
688    in the \fBpcrepattern\fP documentation. Note that the end of this type of
689    comment is a literal newline sequence in the pattern; escape sequences that
690    happen to represent a newline do not count.
691    .P
692  This option makes it possible to include comments inside complicated patterns.  This option makes it possible to include comments inside complicated patterns.
693  Note, however, that this applies only to data characters. Whitespace characters  Note, however, that this applies only to data characters. White space characters
694  may never appear within special character sequences in a pattern, for example  may never appear within special character sequences in a pattern, for example
695  within the sequence (?( which introduces a conditional subpattern.  within the sequence (?( that introduces a conditional subpattern.
696  .sp  .sp
697    PCRE_EXTRA    PCRE_EXTRA
698  .sp  .sp
# Line 548  set, any backslash in a pattern that is Line 702  set, any backslash in a pattern that is
702  special meaning causes an error, thus reserving these combinations for future  special meaning causes an error, thus reserving these combinations for future
703  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
704  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
705  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
706  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)
707    option setting within a pattern.
708  .sp  .sp
709    PCRE_FIRSTLINE    PCRE_FIRSTLINE
710  .sp  .sp
# Line 570  character). Thus, the pattern AB]CD beco Line 725  character). Thus, the pattern AB]CD beco
725  string (by default this causes the current matching alternative to fail). A  string (by default this causes the current matching alternative to fail). A
726  pattern such as (\e1)(a) succeeds when this option is set (assuming it can find  pattern such as (\e1)(a) succeeds when this option is set (assuming it can find
727  an "a" in the subject), whereas it fails by default, for Perl compatibility.  an "a" in the subject), whereas it fails by default, for Perl compatibility.
728    .P
729    (3) \eU matches an upper case "U" character; by default \eU causes a compile
730    time error (Perl uses \eU to upper case subsequent characters).
731    .P
732    (4) \eu matches a lower case "u" character unless it is followed by four
733    hexadecimal digits, in which case the hexadecimal number defines the code point
734    to match. By default, \eu causes a compile time error (Perl uses it to upper
735    case the following character).
736    .P
737    (5) \ex matches a lower case "x" character unless it is followed by two
738    hexadecimal digits, in which case the hexadecimal number defines the code point
739    to match. By default, as in Perl, a hexadecimal number is always expected after
740    \ex, but it may have zero, one, or two digits (so, for example, \exz matches a
741    binary zero character followed by z).
742  .sp  .sp
743    PCRE_MULTILINE    PCRE_MULTILINE
744  .sp  .sp
# Line 599  indicated by a single character (CR or L Line 768  indicated by a single character (CR or L
768  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
769  CRLF sequence. Setting PCRE_NEWLINE_ANYCRLF specifies that any of the three  CRLF sequence. Setting PCRE_NEWLINE_ANYCRLF specifies that any of the three
770  preceding sequences should be recognized. Setting PCRE_NEWLINE_ANY specifies  preceding sequences should be recognized. Setting PCRE_NEWLINE_ANY specifies
771  that any Unicode newline sequence should be recognized. The Unicode newline  that any Unicode newline sequence should be recognized.
772  sequences are the three just mentioned, plus the single characters VT (vertical  .P
773  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
774  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
775    feed, U+000C), NEL (next line, U+0085), LS (line separator, U+2028), and PS
776    (paragraph separator, U+2029). For the 8-bit library, the last two are
777  recognized only in UTF-8 mode.  recognized only in UTF-8 mode.
778  .P  .P
779    When PCRE is compiled to run in an EBCDIC (mainframe) environment, the code for
780    CR is 0x0d, the same as ASCII. However, the character code for LF is normally
781    0x15, though in some EBCDIC environments 0x25 is used. Whichever of these is
782    not LF is made to correspond to Unicode's NEL character. EBCDIC codes are all
783    less than 256. For more details, see the
784    .\" HREF
785    \fBpcrebuild\fP
786    .\"
787    documentation.
788    .P
789  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
790  as a number, giving eight possibilities. Currently only six are used (default  as a number, giving eight possibilities. Currently only six are used (default
791  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 612  option, the combination may or may not b Line 793  option, the combination may or may not b
793  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
794  other combinations may yield unused numbers and cause an error.  other combinations may yield unused numbers and cause an error.
795  .P  .P
796  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
797  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,
798  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
799  line break sequence. In other circumstances, line break sequences are treated  indicates a comment that lasts until after the next line break sequence. In
800  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
801  as whitespace characters and are therefore ignored.  data.
802  .P  .P
803  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
804  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 630  were followed by ?: but named parenthese Line 811  were followed by ?: but named parenthese
811  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
812  in Perl.  in Perl.
813  .sp  .sp
814      NO_START_OPTIMIZE
815    .sp
816    This is an option that acts at matching time; that is, it is really an option
817    for \fBpcre_exec()\fP or \fBpcre_dfa_exec()\fP. If it is set at compile time,
818    it is remembered with the compiled pattern and assumed at matching time. For
819    details see the discussion of PCRE_NO_START_OPTIMIZE
820    .\" HTML <a href="#execoptions">
821    .\" </a>
822    below.
823    .\"
824    .sp
825      PCRE_UCP
826    .sp
827    This option changes the way PCRE processes \eB, \eb, \eD, \ed, \eS, \es, \eW,
828    \ew, and some of the POSIX character classes. By default, only ASCII characters
829    are recognized, but if PCRE_UCP is set, Unicode properties are used instead to
830    classify characters. More details are given in the section on
831    .\" HTML <a href="pcre.html#genericchartypes">
832    .\" </a>
833    generic character types
834    .\"
835    in the
836    .\" HREF
837    \fBpcrepattern\fP
838    .\"
839    page. If you set PCRE_UCP, matching one of the items it affects takes much
840    longer. The option is available only if PCRE has been compiled with Unicode
841    property support.
842    .sp
843    PCRE_UNGREEDY    PCRE_UNGREEDY
844  .sp  .sp
845  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 639  with Perl. It can also be set by a (?U) Line 849  with Perl. It can also be set by a (?U)
849    PCRE_UTF8    PCRE_UTF8
850  .sp  .sp
851  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
852  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
853  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
854  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
855  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  
856  .\" HREF  .\" HREF
857  \fBpcre\fP  \fBpcreunicode\fP
858  .\"  .\"
859  page.  page.
860  .sp  .sp
861    PCRE_NO_UTF8_CHECK    PCRE_NO_UTF8_CHECK
862  .sp  .sp
863  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
864  automatically checked. There is a discussion about the  string is automatically checked. There is a discussion about the
865  .\" HTML <a href="pcre.html#utf8strings">  .\" HTML <a href="pcreunicode.html#utf8strings">
866  .\" </a>  .\" </a>
867  validity of UTF-8 strings  validity of UTF-8 strings
868  .\"  .\"
869  in the main  in the
870  .\" HREF  .\" HREF
871  \fBpcre\fP  \fBpcreunicode\fP
872  .\"  .\"
873  page. If an invalid UTF-8 sequence of bytes is found, \fBpcre_compile()\fP  page. If an invalid UTF-8 sequence is found, \fBpcre_compile()\fP returns an
874  returns an error. If you already know that your pattern is valid, and you want  error. If you already know that your pattern is valid, and you want to skip
875  to skip this check for performance reasons, you can set the PCRE_NO_UTF8_CHECK  this check for performance reasons, you can set the PCRE_NO_UTF8_CHECK option.
876  option. When it is set, the effect of passing an invalid UTF-8 string as a  When it is set, the effect of passing an invalid UTF-8 string as a pattern is
877  pattern is undefined. It may cause your program to crash. Note that this option  undefined. It may cause your program to crash. Note that this option can also
878  can also be passed to \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP, to suppress  be passed to \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP, to suppress the
879  the UTF-8 validity checking of subject strings.  validity checking of subject strings.
880  .  .
881  .  .
882  .SH "COMPILATION ERROR CODES"  .SH "COMPILATION ERROR CODES"
# Line 679  the UTF-8 validity checking of subject s Line 884  the UTF-8 validity checking of subject s
884  .sp  .sp
885  The following table lists the error codes than may be returned by  The following table lists the error codes than may be returned by
886  \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
887  both compiling functions. As PCRE has developed, some error codes have fallen  both compiling functions. Note that error messages are always 8-bit ASCII
888  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
889    have fallen out of use. To avoid confusion, they have not been re-used.
890  .sp  .sp
891     0  no error     0  no error
892     1  \e at end of pattern     1  \e at end of pattern
# Line 714  out of use. To avoid confusion, they hav Line 920  out of use. To avoid confusion, they hav
920    29  (?R or (?[+-]digits must be followed by )    29  (?R or (?[+-]digits must be followed by )
921    30  unknown POSIX class name    30  unknown POSIX class name
922    31  POSIX collating elements are not supported    31  POSIX collating elements are not supported
923    32  this version of PCRE is not compiled with PCRE_UTF8 support    32  this version of PCRE is compiled without UTF support
924    33  [this code is not in use]    33  [this code is not in use]
925    34  character value in \ex{...} sequence is too large    34  character value in \ex{...} sequence is too large
926    35  invalid condition (?(0)    35  invalid condition (?(0)
927    36  \eC not allowed in lookbehind assertion    36  \eC not allowed in lookbehind assertion
928    37  PCRE does not support \eL, \el, \eN, \eU, or \eu    37  PCRE does not support \eL, \el, \eN{name}, \eU, or \eu
929    38  number after (?C is > 255    38  number after (?C is > 255
930    39  closing ) for (?C expected    39  closing ) for (?C expected
931    40  recursive call could loop indefinitely    40  recursive call could loop indefinitely
932    41  unrecognized character after (?P    41  unrecognized character after (?P
933    42  syntax error in subpattern name (missing terminator)    42  syntax error in subpattern name (missing terminator)
934    43  two named subpatterns have the same name    43  two named subpatterns have the same name
935    44  invalid UTF-8 string    44  invalid UTF-8 string (specifically UTF-8)
936    45  support for \eP, \ep, and \eX has not been compiled    45  support for \eP, \ep, and \eX has not been compiled
937    46  malformed \eP or \ep sequence    46  malformed \eP or \ep sequence
938    47  unknown property name after \eP or \ep    47  unknown property name after \eP or \ep
939    48  subpattern name is too long (maximum 32 characters)    48  subpattern name is too long (maximum 32 characters)
940    49  too many named subpatterns (maximum 10000)    49  too many named subpatterns (maximum 10000)
941    50  [this code is not in use]    50  [this code is not in use]
942    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
943    52  internal error: overran compiling workspace    52  internal error: overran compiling workspace
944    53  internal error: previously-checked referenced subpattern not found    53  internal error: previously-checked referenced subpattern
945            not found
946    54  DEFINE group contains more than one branch    54  DEFINE group contains more than one branch
947    55  repeating a DEFINE group is not allowed    55  repeating a DEFINE group is not allowed
948    56  inconsistent NEWLINE options    56  inconsistent NEWLINE options
949    57  \eg is not followed by a braced, angle-bracketed, or quoted    57  \eg is not followed by a braced, angle-bracketed, or quoted
950          name/number or by a plain number          name/number or by a plain number
951    58  a numbered reference must not be zero    58  a numbered reference must not be zero
952    59  (*VERB) with an argument is not supported    59  an argument is not allowed for (*ACCEPT), (*FAIL), or (*COMMIT)
953    60  (*VERB) not recognized    60  (*VERB) not recognized
954    61  number is too big    61  number is too big
955    62  subpattern name expected    62  subpattern name expected
956    63  digit expected after (?+    63  digit expected after (?+
957    64  ] is an invalid data character in JavaScript compatibility mode    64  ] is an invalid data character in JavaScript compatibility mode
958      65  different names for subpatterns of the same number are
959            not allowed
960      66  (*MARK) must have an argument
961      67  this version of PCRE is not compiled with Unicode property
962            support
963      68  \ec must be followed by an ASCII character
964      69  \ek is not followed by a braced, angle-bracketed, or quoted name
965      70  internal error: unknown opcode in find_fixedlength()
966      71  \eN is not supported in a class
967      72  too many forward references
968      73  disallowed Unicode code point (>= 0xd800 && <= 0xdfff)
969      74  invalid UTF-16 string (specifically UTF-16)
970      75  name is too long in (*MARK), (*PRUNE), (*SKIP), or (*THEN)
971      76  character value in \eu.... sequence is too large
972      77  invalid UTF-32 string (specifically UTF-32)
973  .sp  .sp
974  The numbers 32 and 10000 in errors 48 and 49 are defaults; different values may  The numbers 32 and 10000 in errors 48 and 49 are defaults; different values may
975  be used if the limits were changed when PCRE was built.  be used if the limits were changed when PCRE was built.
976  .  .
977  .  .
978    .\" HTML <a name="studyingapattern"></a>
979  .SH "STUDYING A PATTERN"  .SH "STUDYING A PATTERN"
980  .rs  .rs
981  .sp  .sp
# Line 769  help speed up matching, \fBpcre_study()\ Line 992  help speed up matching, \fBpcre_study()\
992  results of the study.  results of the study.
993  .P  .P
994  The returned value from \fBpcre_study()\fP can be passed directly to  The returned value from \fBpcre_study()\fP can be passed directly to
995  \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
996  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
997  described  passed; these are described
998  .\" HTML <a href="#extradata">  .\" HTML <a href="#extradata">
999  .\" </a>  .\" </a>
1000  below  below
1001  .\"  .\"
1002  in the section on matching a pattern.  in the section on matching a pattern.
1003  .P  .P
1004  If studying the pattern does not produce any additional information  If studying the pattern does not produce any useful information,
1005  \fBpcre_study()\fP returns NULL. In that circumstance, if the calling program  \fBpcre_study()\fP returns NULL by default. In that circumstance, if the
1006  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
1007  own \fBpcre_extra\fP block.  \fBpcre_dfa_exec()\fP, it must set up its own \fBpcre_extra\fP block. However,
1008  .P  if \fBpcre_study()\fP is called with the PCRE_STUDY_EXTRA_NEEDED option, it
1009  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
1010  options are defined, and this argument should always be zero.  information. It may still return NULL, however, if an error occurs in
1011    \fBpcre_study()\fP.
1012    .P
1013    The second argument of \fBpcre_study()\fP contains option bits. There are three
1014    further options in addition to PCRE_STUDY_EXTRA_NEEDED:
1015    .sp
1016      PCRE_STUDY_JIT_COMPILE
1017      PCRE_STUDY_JIT_PARTIAL_HARD_COMPILE
1018      PCRE_STUDY_JIT_PARTIAL_SOFT_COMPILE
1019    .sp
1020    If any of these are set, and the just-in-time compiler is available, the
1021    pattern is further compiled into machine code that executes much faster than
1022    the \fBpcre_exec()\fP interpretive matching function. If the just-in-time
1023    compiler is not available, these options are ignored. All undefined bits in the
1024    \fIoptions\fP argument must be zero.
1025    .P
1026    JIT compilation is a heavyweight optimization. It can take some time for
1027    patterns to be analyzed, and for one-off matches and simple patterns the
1028    benefit of faster execution might be offset by a much slower study time.
1029    Not all patterns can be optimized by the JIT compiler. For those that cannot be
1030    handled, matching automatically falls back to the \fBpcre_exec()\fP
1031    interpreter. For more details, see the
1032    .\" HREF
1033    \fBpcrejit\fP
1034    .\"
1035    documentation.
1036  .P  .P
1037  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
1038  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 793  static string that is part of the librar Line 1041  static string that is part of the librar
1041  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
1042  sure that it has run successfully.  sure that it has run successfully.
1043  .P  .P
1044  This is a typical call to \fBpcre_study\fP():  When you are finished with a pattern, you can free the memory used for the
1045    study data by calling \fBpcre_free_study()\fP. This function was added to the
1046    API for release 8.20. For earlier versions, the memory could be freed with
1047    \fBpcre_free()\fP, just like the pattern itself. This will still work in cases
1048    where JIT optimization is not used, but it is advisable to change to the new
1049    function when convenient.
1050    .P
1051    This is a typical way in which \fBpcre_study\fP() is used (except that in a
1052    real application there should be tests for errors):
1053  .sp  .sp
1054    pcre_extra *pe;    int rc;
1055    pe = pcre_study(    pcre *re;
1056      pcre_extra *sd;
1057      re = pcre_compile("pattern", 0, &error, &erroroffset, NULL);
1058      sd = pcre_study(
1059      re,             /* result of pcre_compile() */      re,             /* result of pcre_compile() */
1060      0,              /* no options exist */      0,              /* no options */
1061      &error);        /* set to NULL or points to a message */      &error);        /* set to NULL or points to a message */
1062  .sp    rc = pcre_exec(   /* see below for details of pcre_exec() options */
1063  At present, studying a pattern is useful only for non-anchored patterns that do      re, sd, "subject", 7, 0, 0, ovector, 30);
1064  not have a single fixed starting character. A bitmap of possible starting    ...
1065  bytes is created.    pcre_free_study(sd);
1066      pcre_free(re);
1067    .sp
1068    Studying a pattern does two things: first, a lower bound for the length of
1069    subject string that is needed to match the pattern is computed. This does not
1070    mean that there are any strings of that length that match, but it does
1071    guarantee that no shorter strings match. The value is used to avoid wasting
1072    time by trying to match strings that are shorter than the lower bound. You can
1073    find out the value in a calling program via the \fBpcre_fullinfo()\fP function.
1074    .P
1075    Studying a pattern is also useful for non-anchored patterns that do not have a
1076    single fixed starting character. A bitmap of possible starting bytes is
1077    created. This speeds up finding a position in the subject at which to start
1078    matching. (In 16-bit mode, the bitmap is used for 16-bit values less than 256.
1079    In 32-bit mode, the bitmap is used for 32-bit values less than 256.)
1080    .P
1081    These two optimizations apply to both \fBpcre_exec()\fP and
1082    \fBpcre_dfa_exec()\fP, and the information is also used by the JIT compiler.
1083    The optimizations can be disabled by setting the PCRE_NO_START_OPTIMIZE option
1084    when calling \fBpcre_exec()\fP or \fBpcre_dfa_exec()\fP, but if this is done,
1085    JIT execution is also disabled. You might want to do this if your pattern
1086    contains callouts or (*MARK) and you want to make use of these facilities in
1087    cases where matching fails. See the discussion of PCRE_NO_START_OPTIMIZE
1088    .\" HTML <a href="#execoptions">
1089    .\" </a>
1090    below.
1091    .\"
1092  .  .
1093  .  .
1094  .\" HTML <a name="localesupport"></a>  .\" HTML <a name="localesupport"></a>
# Line 812  bytes is created. Line 1097  bytes is created.
1097  .sp  .sp
1098  PCRE handles caseless matching, and determines whether characters are letters,  PCRE handles caseless matching, and determines whether characters are letters,
1099  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
1100  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
1101  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
1102  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
1103  support. The use of locales with Unicode is discouraged. If you are handling  Unicode character property support. Alternatively, the PCRE_UCP option can be
1104  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
1105  Unicode, or use locales, but not try to mix the two.  support instead of built-in tables. The use of locales with Unicode is
1106    discouraged. If you are handling characters with codes greater than 128, you
1107    should either use UTF-8 and Unicode, or use locales, but not try to mix the
1108    two.
1109  .P  .P
1110  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
1111  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 862  one in which it was compiled. Passing ta Line 1150  one in which it was compiled. Passing ta
1150  below in the section on matching a pattern.  below in the section on matching a pattern.
1151  .  .
1152  .  .
1153    .\" HTML <a name="infoaboutpattern"></a>
1154  .SH "INFORMATION ABOUT A PATTERN"  .SH "INFORMATION ABOUT A PATTERN"
1155  .rs  .rs
1156  .sp  .sp
# Line 870  below in the section on matching a patte Line 1159  below in the section on matching a patte
1159  .B int \fIwhat\fP, void *\fIwhere\fP);  .B int \fIwhat\fP, void *\fIwhere\fP);
1160  .PP  .PP
1161  The \fBpcre_fullinfo()\fP function returns information about a compiled  The \fBpcre_fullinfo()\fP function returns information about a compiled
1162  pattern. It replaces the obsolete \fBpcre_info()\fP function, which is  pattern. It replaces the \fBpcre_info()\fP function, which was removed from the
1163  nevertheless retained for backwards compability (and is documented below).  library at version 8.30, after more than 10 years of obsolescence.
1164  .P  .P
1165  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
1166  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 880  information is required, and the fourth Line 1169  information is required, and the fourth
1169  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
1170  the following negative numbers:  the following negative numbers:
1171  .sp  .sp
1172    PCRE_ERROR_NULL       the argument \fIcode\fP was NULL    PCRE_ERROR_NULL           the argument \fIcode\fP was NULL
1173                          the argument \fIwhere\fP was NULL                              the argument \fIwhere\fP was NULL
1174    PCRE_ERROR_BADMAGIC   the "magic number" was not found    PCRE_ERROR_BADMAGIC       the "magic number" was not found
1175    PCRE_ERROR_BADOPTION  the value of \fIwhat\fP was invalid    PCRE_ERROR_BADENDIANNESS  the pattern was compiled with different
1176                                endianness
1177      PCRE_ERROR_BADOPTION      the value of \fIwhat\fP was invalid
1178  .sp  .sp
1179  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
1180  check against passing an arbitrary memory pointer. Here is a typical call of  check against passing an arbitrary memory pointer. The endianness error can
1181  \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
1182    a typical call of \fBpcre_fullinfo()\fP, to obtain the length of the compiled
1183    pattern:
1184  .sp  .sp
1185    int rc;    int rc;
1186    size_t length;    size_t length;
1187    rc = pcre_fullinfo(    rc = pcre_fullinfo(
1188      re,               /* result of pcre_compile() */      re,               /* result of pcre_compile() */
1189      pe,               /* result of pcre_study(), or NULL */      sd,               /* result of pcre_study(), or NULL */
1190      PCRE_INFO_SIZE,   /* what is required */      PCRE_INFO_SIZE,   /* what is required */
1191      &length);         /* where to put the data */      &length);         /* where to put the data */
1192  .sp  .sp
# Line 921  a NULL table pointer. Line 1214  a NULL table pointer.
1214  .sp  .sp
1215    PCRE_INFO_FIRSTBYTE    PCRE_INFO_FIRSTBYTE
1216  .sp  .sp
1217  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
1218  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,
1219  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
1220  still recognized for backwards compatibility.)  variable.
1221    .P
1222    If there is a fixed first value, for example, the letter "c" from a pattern
1223    such as (cat|cow|coyote), its value is returned. In the 8-bit library, the
1224    value is always less than 256. In the 16-bit library the value can be up to
1225    0xffff. In the 32-bit library the value can be up to 0x10ffff.
1226  .P  .P
1227  If there is a fixed first byte, for example, from a pattern such as  If there is no fixed first value, and if either
 (cat|cow|coyote), its value is returned. Otherwise, if either  
1228  .sp  .sp
1229  (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
1230  starts with "^", or  starts with "^", or
# Line 938  starts with "^", or Line 1235  starts with "^", or
1235  -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
1236  subject string or after any newline within the string. Otherwise -2 is  subject string or after any newline within the string. Otherwise -2 is
1237  returned. For anchored patterns, -2 is returned.  returned. For anchored patterns, -2 is returned.
1238    .P
1239    Since for the 32-bit library using the non-UTF-32 mode, this function is unable
1240    to return the full 32-bit range of the character, this value is deprecated;
1241    instead the PCRE_INFO_FIRSTLITERALSET and PCRE_INFO_FIRSTLITERAL values should
1242    be used.
1243  .sp  .sp
1244    PCRE_INFO_FIRSTTABLE    PCRE_INFO_FIRSTTABLE
1245  .sp  .sp
1246  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
1247  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
1248  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
1249  fourth argument should point to an \fBunsigned char *\fP variable.  fourth argument should point to an \fBunsigned char *\fP variable.
1250  .sp  .sp
# Line 958  Return 1 if the (?J) or (?-J) option set Line 1260  Return 1 if the (?J) or (?-J) option set
1260  0. The fourth argument should point to an \fBint\fP variable. (?J) and  0. The fourth argument should point to an \fBint\fP variable. (?J) and
1261  (?-J) set and unset the local PCRE_DUPNAMES option, respectively.  (?-J) set and unset the local PCRE_DUPNAMES option, respectively.
1262  .sp  .sp
1263      PCRE_INFO_JIT
1264    .sp
1265    Return 1 if the pattern was studied with one of the JIT options, and
1266    just-in-time compiling was successful. The fourth argument should point to an
1267    \fBint\fP variable. A return value of 0 means that JIT support is not available
1268    in this version of PCRE, or that the pattern was not studied with a JIT option,
1269    or that the JIT compiler could not handle this particular pattern. See the
1270    .\" HREF
1271    \fBpcrejit\fP
1272    .\"
1273    documentation for details of what can and cannot be handled.
1274    .sp
1275      PCRE_INFO_JITSIZE
1276    .sp
1277    If the pattern was successfully studied with a JIT option, return the size of
1278    the JIT compiled code, otherwise return zero. The fourth argument should point
1279    to a \fBsize_t\fP variable.
1280    .sp
1281    PCRE_INFO_LASTLITERAL    PCRE_INFO_LASTLITERAL
1282  .sp  .sp
1283  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
1284  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
1285  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
1286  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
1287  follows something of variable length. For example, for the pattern  only if it follows something of variable length. For example, for the pattern
1288  /^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
1289  is -1.  is -1.
1290    .P
1291    Since for the 32-bit library using the non-UTF-32 mode, this function is unable
1292    to return the full 32-bit range of the character, this value is deprecated;
1293    instead the PCRE_INFO_LASTLITERAL2SET and PCRE_INFO_LASTLITERAL2 values should
1294    be used.
1295    .sp
1296      PCRE_INFO_MAXLOOKBEHIND
1297    .sp
1298    Return the number of characters (NB not bytes) in the longest lookbehind
1299    assertion in the pattern. Note that the simple assertions \eb and \eB require a
1300    one-character lookbehind. This information is useful when doing multi-segment
1301    matching using the partial matching facilities.
1302    .sp
1303      PCRE_INFO_MINLENGTH
1304    .sp
1305    If the pattern was studied and a minimum length for matching subject strings
1306    was computed, its value is returned. Otherwise the returned value is -1. The
1307    value is a number of characters, which in UTF-8 mode may be different from the
1308    number of bytes. The fourth argument should point to an \fBint\fP variable. A
1309    non-negative value is a lower bound to the length of any matching string. There
1310    may not be any strings of that length that do actually match, but every string
1311    that does match is at least that long.
1312  .sp  .sp
1313    PCRE_INFO_NAMECOUNT    PCRE_INFO_NAMECOUNT
1314    PCRE_INFO_NAMEENTRYSIZE    PCRE_INFO_NAMEENTRYSIZE
# Line 986  The map consists of a number of fixed-si Line 1328  The map consists of a number of fixed-si
1328  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
1329  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
1330  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
1331  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
1332  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,
1333  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
1334  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.
1335  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
1336  PCRE_EXTENDED is set, so white space - including newlines - is ignored):  which contains the parenthesis number. The rest
1337    of the entry is the corresponding name, zero terminated.
1338    .P
1339    The names are in alphabetical order. Duplicate names may appear if (?| is used
1340    to create multiple groups with the same number, as described in the
1341    .\" HTML <a href="pcrepattern.html#dupsubpatternnumber">
1342    .\" </a>
1343    section on duplicate subpattern numbers
1344    .\"
1345    in the
1346    .\" HREF
1347    \fBpcrepattern\fP
1348    .\"
1349    page. Duplicate names for subpatterns with different numbers are permitted only
1350    if PCRE_DUPNAMES is set. In all cases of duplicate names, they appear in the
1351    table in the order in which they were found in the pattern. In the absence of
1352    (?| this is the order of increasing number; when (?| is used this is not
1353    necessarily the case because later subpatterns may have lower numbers.
1354    .P
1355    As a simple example of the name/number table, consider the following pattern
1356    after compilation by the 8-bit library (assume PCRE_EXTENDED is set, so white
1357    space - including newlines - is ignored):
1358  .sp  .sp
1359  .\" JOIN  .\" JOIN
1360    (?<date> (?<year>(\ed\ed)?\ed\ed) -    (?<date> (?<year>(\ed\ed)?\ed\ed) -
# Line 1046  For such patterns, the PCRE_ANCHORED bit Line 1409  For such patterns, the PCRE_ANCHORED bit
1409  .sp  .sp
1410    PCRE_INFO_SIZE    PCRE_INFO_SIZE
1411  .sp  .sp
1412  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
1413  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
1414  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
1415  variable.  \fBpcre_compile()\fP. The value that is passed as the argument to
1416    \fBpcre_malloc()\fP when \fBpcre_compile()\fP is getting memory in which to
1417    place the compiled data is the value returned by this option plus the size of
1418    the \fBpcre\fP structure. Studying a compiled pattern, with or without JIT,
1419    does not alter the value returned by this option.
1420  .sp  .sp
1421    PCRE_INFO_STUDYSIZE    PCRE_INFO_STUDYSIZE
1422  .sp  .sp
1423  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
1424  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
1425  \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
1426  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
1427  \fBsize_t\fP variable.  to record information that will speed up matching (see the section entitled
1428  .  .\" HTML <a href="#studyingapattern">
1429  .  .\" </a>
1430  .SH "OBSOLETE INFO FUNCTION"  "Studying a pattern"
1431  .rs  .\"
1432    above). The format of the \fIstudy_data\fP block is private, but its length
1433    is made available via this option so that it can be saved and restored (see the
1434    .\" HREF
1435    \fBpcreprecompile\fP
1436    .\"
1437    documentation for details).
1438  .sp  .sp
1439  .B int pcre_info(const pcre *\fIcode\fP, int *\fIoptptr\fP, int    PCRE_INFO_FIRSTLITERALSET
1440  .B *\fIfirstcharptr\fP);  .sp
1441  .PP  Return information about the first data unit of any matched string, for a
1442  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
1443  restrictive to return all the available data about a compiled pattern. New  variable.
1444  programs should use \fBpcre_fullinfo()\fP instead. The yield of  .P
1445  \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
1446  following negative numbers:  such as (cat|cow|coyote), 1 is returned, and the character value can be
1447    retrieved using PCRE_INFO_FIRSTLITERAL.
1448    .P
1449    If there is no fixed first value, and if either
1450    .sp
1451    (a) the pattern was compiled with the PCRE_MULTILINE option, and every branch
1452    starts with "^", or
1453  .sp  .sp
1454    PCRE_ERROR_NULL       the argument \fIcode\fP was NULL  (b) every branch of the pattern starts with ".*" and PCRE_DOTALL is not set
1455    PCRE_ERROR_BADMAGIC   the "magic number" was not found  (if it were set, the pattern would be anchored),
1456  .sp  .sp
1457  If the \fIoptptr\fP argument is not NULL, a copy of the options with which the  2 is returned, indicating that the pattern matches only at the start of a
1458  pattern was compiled is placed in the integer it points to (see  subject string or after any newline within the string. Otherwise 0 is
1459  PCRE_INFO_OPTIONS above).  returned. For anchored patterns, 0 is returned.
1460    .sp
1461      PCRE_INFO_FIRSTLITERAL
1462    .sp
1463    Return the fixed first character value, if PCRE_INFO_FIRSTLITERALSET returned 1;
1464    otherwise returns 0. The fourth argument should point to an \fBuint_t\fP
1465    variable.
1466  .P  .P
1467  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
1468  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
1469  string (see PCRE_INFO_FIRSTBYTE above).  can be up to 0x10ffff, and up to 0xffffffff when not using UTF-32 mode.
1470    .P
1471    If there is no fixed first value, and if either
1472    .sp
1473    (a) the pattern was compiled with the PCRE_MULTILINE option, and every branch
1474    starts with "^", or
1475    .sp
1476    (b) every branch of the pattern starts with ".*" and PCRE_DOTALL is not set
1477    (if it were set, the pattern would be anchored),
1478    .sp
1479    -1 is returned, indicating that the pattern matches only at the start of a
1480    subject string or after any newline within the string. Otherwise -2 is
1481    returned. For anchored patterns, -2 is returned.
1482    .sp
1483      PCRE_INFO_LASTLITERAL2SET
1484    .sp
1485    Returns 1 if there is a rightmost literal data unit that must exist in any matched
1486    string, other than at its start. The fourth argument should  point to an \fBint\fP
1487    variable. If there is no such value, 0 is returned. If returning 1, the character
1488    value itself can be retrieved using PCRE_INFO_LASTLITERAL2.
1489    .P
1490    For anchored patterns, a last literal value is recorded only if it follows something
1491    of variable length. For example, for the pattern /^a\ed+z\ed+/ the returned value
1492    1 (with "z" returned from PCRE_INFO_LASTLITERAL2), but for /^a\edz\ed/ the returned
1493    value is 0.
1494    .sp
1495      PCRE_INFO_LASTLITERAL2
1496    .sp
1497    Return the value of the rightmost literal data unit that must exist in any
1498    matched string, other than at its start, if such a value has been recorded. The
1499    fourth argument should point to an \fBuint32_t\fP variable. If there is no such
1500    value, 0 is returned.
1501  .  .
1502  .  .
1503  .SH "REFERENCE COUNTS"  .SH "REFERENCE COUNTS"
# Line 1118  is different. (This seems a highly unlik Line 1534  is different. (This seems a highly unlik
1534  .P  .P
1535  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
1536  compiled pattern, which is passed in the \fIcode\fP argument. If the  compiled pattern, which is passed in the \fIcode\fP argument. If the
1537  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
1538  \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
1539  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
1540  also an alternative matching function, which is described  different subject strings with the same pattern.
1541    .P
1542    This function is the main matching facility of the library, and it operates in
1543    a Perl-like manner. For specialist use there is also an alternative matching
1544    function, which is described
1545  .\" HTML <a href="#dfamatch">  .\" HTML <a href="#dfamatch">
1546  .\" </a>  .\" </a>
1547  below  below
# Line 1152  Here is an example of a simple call to \ Line 1572  Here is an example of a simple call to \
1572      ovector,        /* vector of integers for substring information */      ovector,        /* vector of integers for substring information */
1573      30);            /* number of elements (NOT size in bytes) */      30);            /* number of elements (NOT size in bytes) */
1574  .  .
1575    .
1576  .\" HTML <a name="extradata"></a>  .\" HTML <a name="extradata"></a>
1577  .SS "Extra data for \fBpcre_exec()\fR"  .SS "Extra data for \fBpcre_exec()\fR"
1578  .rs  .rs
# Line 1164  fields (not necessarily in this order): Line 1585  fields (not necessarily in this order):
1585  .sp  .sp
1586    unsigned long int \fIflags\fP;    unsigned long int \fIflags\fP;
1587    void *\fIstudy_data\fP;    void *\fIstudy_data\fP;
1588      void *\fIexecutable_jit\fP;
1589    unsigned long int \fImatch_limit\fP;    unsigned long int \fImatch_limit\fP;
1590    unsigned long int \fImatch_limit_recursion\fP;    unsigned long int \fImatch_limit_recursion\fP;
1591    void *\fIcallout_data\fP;    void *\fIcallout_data\fP;
1592    const unsigned char *\fItables\fP;    const unsigned char *\fItables\fP;
1593      unsigned char **\fImark\fP;
1594  .sp  .sp
1595  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
1596  are set. The flag bits are:  "PCRE_UCHAR16 **".
1597  .sp  .sp
1598    PCRE_EXTRA_STUDY_DATA  In the 32-bit version of this structure, the \fImark\fP field has type
1599    "PCRE_UCHAR32 **".
1600    .P
1601    The \fIflags\fP field is used to specify which of the other fields are set. The
1602    flag bits are:
1603    .sp
1604      PCRE_EXTRA_CALLOUT_DATA
1605      PCRE_EXTRA_EXECUTABLE_JIT
1606      PCRE_EXTRA_MARK
1607    PCRE_EXTRA_MATCH_LIMIT    PCRE_EXTRA_MATCH_LIMIT
1608    PCRE_EXTRA_MATCH_LIMIT_RECURSION    PCRE_EXTRA_MATCH_LIMIT_RECURSION
1609    PCRE_EXTRA_CALLOUT_DATA    PCRE_EXTRA_STUDY_DATA
1610    PCRE_EXTRA_TABLES    PCRE_EXTRA_TABLES
1611  .sp  .sp
1612  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
1613  \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
1614  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
1615  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
1616    fields and their corresponding flag bits.
1617  .P  .P
1618  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
1619  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,
1620  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
1621  classic example is the use of nested unlimited repeats.  classic example is a pattern that uses nested unlimited repeats.
1622  .P  .P
1623  Internally, PCRE uses a function called \fBmatch()\fP which it calls repeatedly  Internally, \fBpcre_exec()\fP uses a function called \fBmatch()\fP, which it
1624  (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
1625  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
1626  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
1627  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
1628  string.  in the subject string.
1629    .P
1630    When \fBpcre_exec()\fP is called with a pattern that was successfully studied
1631    with a JIT option, the way that the matching is executed is entirely different.
1632    However, there is still the possibility of runaway matching that goes on for a
1633    very long time, and so the \fImatch_limit\fP value is also used in this case
1634    (but in a different way) to limit how long the matching can continue.
1635  .P  .P
1636  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
1637  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 1208  limits the depth of recursion. The recur Line 1646  limits the depth of recursion. The recur
1646  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.
1647  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.
1648  .P  .P
1649  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
1650  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
1651  amount of heap memory that can be used.  stack, the amount of heap memory that can be used. This limit is not relevant,
1652    and is ignored, when matching is done using JIT compiled code.
1653  .P  .P
1654  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
1655  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 1239  called. See the Line 1678  called. See the
1678  \fBpcreprecompile\fP  \fBpcreprecompile\fP
1679  .\"  .\"
1680  documentation for a discussion of saving compiled patterns for later use.  documentation for a discussion of saving compiled patterns for later use.
1681    .P
1682    If PCRE_EXTRA_MARK is set in the \fIflags\fP field, the \fImark\fP field must
1683    be set to point to a suitable variable. If the pattern contains any
1684    backtracking control verbs such as (*MARK:NAME), and the execution ends up with
1685    a name to pass back, a pointer to the name string (zero terminated) is placed
1686    in the variable pointed to by the \fImark\fP field. The names are within the
1687    compiled pattern; if you wish to retain such a name you must copy it before
1688    freeing the memory of a compiled pattern. If there is no name to pass back, the
1689    variable pointed to by the \fImark\fP field is set to NULL. For details of the
1690    backtracking control verbs, see the section entitled
1691    .\" HTML <a href="pcrepattern#backtrackcontrol">
1692    .\" </a>
1693    "Backtracking control"
1694    .\"
1695    in the
1696    .\" HREF
1697    \fBpcrepattern\fP
1698    .\"
1699    documentation.
1700    .
1701  .  .
1702  .\" HTML <a name="execoptions"></a>  .\" HTML <a name="execoptions"></a>
1703  .SS "Option bits for \fBpcre_exec()\fP"  .SS "Option bits for \fBpcre_exec()\fP"
# Line 1247  documentation for a discussion of saving Line 1706  documentation for a discussion of saving
1706  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
1707  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,
1708  PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NOTEMPTY_ATSTART,  PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NOTEMPTY_ATSTART,
1709  PCRE_NO_START_OPTIMIZE, PCRE_NO_UTF8_CHECK, PCRE_PARTIAL_SOFT, and  PCRE_NO_START_OPTIMIZE, PCRE_NO_UTF8_CHECK, PCRE_PARTIAL_HARD, and
1710  PCRE_PARTIAL_HARD.  PCRE_PARTIAL_SOFT.
1711    .P
1712    If the pattern was successfully studied with one of the just-in-time (JIT)
1713    compile options, the only supported options for JIT execution are
1714    PCRE_NO_UTF8_CHECK, PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY,
1715    PCRE_NOTEMPTY_ATSTART, PCRE_PARTIAL_HARD, and PCRE_PARTIAL_SOFT. If an
1716    unsupported option is used, JIT execution is disabled and the normal
1717    interpretive code in \fBpcre_exec()\fP is run.
1718  .sp  .sp
1719    PCRE_ANCHORED    PCRE_ANCHORED
1720  .sp  .sp
# Line 1329  valid, so PCRE searches further into the Line 1795  valid, so PCRE searches further into the
1795  .sp  .sp
1796    PCRE_NOTEMPTY_ATSTART    PCRE_NOTEMPTY_ATSTART
1797  .sp  .sp
1798  This is like PCRE_NOTEMPTY, except that an empty string match that is not at  This is like PCRE_NOTEMPTY, except that an empty string match that is not at
1799  the start of the subject is permitted. If the pattern is anchored, such a match  the start of the subject is permitted. If the pattern is anchored, such a match
1800  can occur only if the pattern contains \eK.  can occur only if the pattern contains \eK.
1801  .P  .P
# Line 1344  the Line 1810  the
1810  .\" HREF  .\" HREF
1811  \fBpcredemo\fP  \fBpcredemo\fP
1812  .\"  .\"
1813  sample program.  sample program. In the most general case, you have to check to see if the
1814    newline convention recognizes CRLF as a newline, and if so, and the current
1815    character is CR followed by LF, advance the starting offset by two characters
1816    instead of one.
1817  .sp  .sp
1818    PCRE_NO_START_OPTIMIZE    PCRE_NO_START_OPTIMIZE
1819  .sp  .sp
1820  There are a number of optimizations that \fBpcre_exec()\fP uses at the start of  There are a number of optimizations that \fBpcre_exec()\fP uses at the start of
1821  a match, in order to speed up the process. For example, if it is known that a  a match, in order to speed up the process. For example, if it is known that an
1822  match must start with a specific character, it searches the subject for that  unanchored match must start with a specific character, it searches the subject
1823  character, and fails immediately if it cannot find it, without actually running  for that character, and fails immediately if it cannot find it, without
1824  the main matching function. When callouts are in use, these optimizations can  actually running the main matching function. This means that a special item
1825  cause them to be skipped. This option disables the "start-up" optimizations,  such as (*COMMIT) at the start of a pattern is not considered until after a
1826  causing performance to suffer, but ensuring that the callouts do occur.  suitable starting point for the match has been found. When callouts or (*MARK)
1827    items are in use, these "start-up" optimizations can cause them to be skipped
1828    if the pattern is never actually used. The start-up optimizations are in effect
1829    a pre-scan of the subject that takes place before the pattern is run.
1830    .P
1831    The PCRE_NO_START_OPTIMIZE option disables the start-up optimizations, possibly
1832    causing performance to suffer, but ensuring that in cases where the result is
1833    "no match", the callouts do occur, and that items such as (*COMMIT) and (*MARK)
1834    are considered at every possible starting position in the subject string. If
1835    PCRE_NO_START_OPTIMIZE is set at compile time, it cannot be unset at matching
1836    time. The use of PCRE_NO_START_OPTIMIZE disables JIT execution; when it is set,
1837    matching is always done using interpretively.
1838    .P
1839    Setting PCRE_NO_START_OPTIMIZE can change the outcome of a matching operation.
1840    Consider the pattern
1841    .sp
1842      (*COMMIT)ABC
1843    .sp
1844    When this is compiled, PCRE records the fact that a match must start with the
1845    character "A". Suppose the subject string is "DEFABC". The start-up
1846    optimization scans along the subject, finds "A" and runs the first match
1847    attempt from there. The (*COMMIT) item means that the pattern must match the
1848    current starting position, which in this case, it does. However, if the same
1849    match is run with PCRE_NO_START_OPTIMIZE set, the initial scan along the
1850    subject string does not happen. The first match attempt is run starting from
1851    "D" and when this fails, (*COMMIT) prevents any further matches being tried, so
1852    the overall result is "no match". If the pattern is studied, more start-up
1853    optimizations may be used. For example, a minimum length for the subject may be
1854    recorded. Consider the pattern
1855    .sp
1856      (*MARK:A)(X|Y)
1857    .sp
1858    The minimum length for a match is one character. If the subject is "ABC", there
1859    will be attempts to match "ABC", "BC", "C", and then finally an empty string.
1860    If the pattern is studied, the final attempt does not take place, because PCRE
1861    knows that the subject is too short, and so the (*MARK) is never encountered.
1862    In this case, studying the pattern does not affect the overall match result,
1863    which is still "no match", but it does affect the auxiliary information that is
1864    returned.
1865  .sp  .sp
1866    PCRE_NO_UTF8_CHECK    PCRE_NO_UTF8_CHECK
1867  .sp  .sp
1868  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
1869  string is automatically checked when \fBpcre_exec()\fP is subsequently called.  string is automatically checked when \fBpcre_exec()\fP is subsequently called.
1870  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
1871  start of a UTF-8 character. There is a discussion about the validity of UTF-8  of \fIstartoffset\fP is also checked to ensure that it points to the start of a
1872  strings in the  UTF-8 character. There is a discussion about the
1873  .\" HTML <a href="pcre.html#utf8strings">  .\" HTML <a href="pcreunicode.html#utf8strings">
1874  .\" </a>  .\" </a>
1875  section on UTF-8 support  validity of UTF-8 strings
1876  .\"  .\"
1877  in the main  in the
1878  .\" HREF  .\" HREF
1879  \fBpcre\fP  \fBpcreunicode\fP
1880  .\"  .\"
1881  page. If an invalid UTF-8 sequence of bytes is found, \fBpcre_exec()\fP returns  page. If an invalid sequence of bytes is found, \fBpcre_exec()\fP returns the
1882  the error PCRE_ERROR_BADUTF8. If \fIstartoffset\fP contains an invalid value,  error PCRE_ERROR_BADUTF8 or, if PCRE_PARTIAL_HARD is set and the problem is a
1883  PCRE_ERROR_BADUTF8_OFFSET is returned.  truncated character at the end of the subject, PCRE_ERROR_SHORTUTF8. In both
1884    cases, information about the precise nature of the error may also be returned
1885    (see the descriptions of these errors in the section entitled \fIError return
1886    values from\fP \fBpcre_exec()\fP
1887    .\" HTML <a href="#errorlist">
1888    .\" </a>
1889    below).
1890    .\"
1891    If \fIstartoffset\fP contains a value that does not point to the start of a
1892    UTF-8 character (or to the end of the subject), PCRE_ERROR_BADUTF8_OFFSET is
1893    returned.
1894  .P  .P
1895  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
1896  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
1897  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
1898  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
1899  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
1900  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
1901  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
1902  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
1903  UTF-8 character, is undefined. Your program may crash.  undefined. Your program may crash.
1904  .sp  .sp
1905    PCRE_PARTIAL_HARD    PCRE_PARTIAL_HARD
1906    PCRE_PARTIAL_SOFT    PCRE_PARTIAL_SOFT
1907  .sp  .sp
1908  These options turn on the partial matching feature. For backwards  These options turn on the partial matching feature. For backwards
1909  compatibility, PCRE_PARTIAL is a synonym for PCRE_PARTIAL_SOFT. A partial match  compatibility, PCRE_PARTIAL is a synonym for PCRE_PARTIAL_SOFT. A partial match
1910  occurs if the end of the subject string is reached successfully, but there are  occurs if the end of the subject string is reached successfully, but there are
1911  not enough subject characters to complete the match. If this happens when  not enough subject characters to complete the match. If this happens when
1912  PCRE_PARTIAL_HARD is set, \fBpcre_exec()\fP immediately returns  PCRE_PARTIAL_SOFT (but not PCRE_PARTIAL_HARD) is set, matching continues by
1913  PCRE_ERROR_PARTIAL. Otherwise, if PCRE_PARTIAL_SOFT is set, matching continues  testing any remaining alternatives. Only if no complete match can be found is
1914  by testing any other alternatives. Only if they all fail is PCRE_ERROR_PARTIAL  PCRE_ERROR_PARTIAL returned instead of PCRE_ERROR_NOMATCH. In other words,
1915  returned (instead of PCRE_ERROR_NOMATCH). The portion of the string that  PCRE_PARTIAL_SOFT says that the caller is prepared to handle a partial match,
1916  was inspected when the partial match was found is set as the first matching  but only if no complete match can be found.
1917  string. There is a more detailed discussion in the  .P
1918    If PCRE_PARTIAL_HARD is set, it overrides PCRE_PARTIAL_SOFT. In this case, if a
1919    partial match is found, \fBpcre_exec()\fP immediately returns
1920    PCRE_ERROR_PARTIAL, without considering any other alternatives. In other words,
1921    when PCRE_PARTIAL_HARD is set, a partial match is considered to be more
1922    important that an alternative complete match.
1923    .P
1924    In both cases, the portion of the string that was inspected when the partial
1925    match was found is set as the first matching string. There is a more detailed
1926    discussion of partial and multi-segment matching, with examples, in the
1927  .\" HREF  .\" HREF
1928  \fBpcrepartial\fP  \fBpcrepartial\fP
1929  .\"  .\"
1930  documentation.  documentation.
1931  .  .
1932    .
1933  .SS "The string to be matched by \fBpcre_exec()\fP"  .SS "The string to be matched by \fBpcre_exec()\fP"
1934  .rs  .rs
1935  .sp  .sp
1936  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
1937  \fIsubject\fP, a length (in bytes) in \fIlength\fP, and a starting byte offset  \fIsubject\fP, a length in bytes in \fIlength\fP, and a starting byte offset
1938  in \fIstartoffset\fP. In UTF-8 mode, the byte offset must point to the start of  in \fIstartoffset\fP. If this is negative or greater than the length of the
1939  a UTF-8 character. Unlike the pattern string, the subject may contain binary  subject, \fBpcre_exec()\fP returns PCRE_ERROR_BADOFFSET. When the starting
1940  zero bytes. When the starting offset is zero, the search for a match starts at  offset is zero, the search for a match starts at the beginning of the subject,
1941  the 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
1942    point to the start of a UTF-8 character (or the end of the subject). Unlike the
1943    pattern string, the subject may contain binary zero bytes.
1944  .P  .P
1945  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
1946  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 1431  start of the subject, which is deemed to Line 1960  start of the subject, which is deemed to
1960  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
1961  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.
1962  .P  .P
1963    Finding all the matches in a subject is tricky when the pattern can match an
1964    empty string. It is possible to emulate Perl's /g behaviour by first trying the
1965    match again at the same offset, with the PCRE_NOTEMPTY_ATSTART and
1966    PCRE_ANCHORED options, and then if that fails, advancing the starting offset
1967    and trying an ordinary match again. There is some code that demonstrates how to
1968    do this in the
1969    .\" HREF
1970    \fBpcredemo\fP
1971    .\"
1972    sample program. In the most general case, you have to check to see if the
1973    newline convention recognizes CRLF as a newline, and if so, and the current
1974    character is CR followed by LF, advance the starting offset by two characters
1975    instead of one.
1976    .P
1977  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
1978  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
1979  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.
1980  .  .
1981    .
1982  .SS "How \fBpcre_exec()\fP returns captured substrings"  .SS "How \fBpcre_exec()\fP returns captured substrings"
1983  .rs  .rs
1984  .sp  .sp
# Line 1478  string that it matched that is returned. Line 2022  string that it matched that is returned.
2022  .P  .P
2023  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
2024  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
2025  returns a value of zero. If the substring offsets are not of interest,  returns a value of zero. If neither the actual string matched nor any captured
2026  \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
2027  \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
2028  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
2029  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
2030  advisable to supply an \fIovector\fP.  is usually advisable to supply an \fIovector\fP of reasonable size.
2031    .P
2032    There are some cases where zero is returned (indicating vector overflow) when
2033    in fact the vector is exactly the right size for the final match. For example,
2034    consider the pattern
2035    .sp
2036      (a)(?:(b)c|bd)
2037    .sp
2038    If a vector of 6 elements (allowing for only 1 captured substring) is given
2039    with subject string "abd", \fBpcre_exec()\fP will try to set the second
2040    captured string, thereby recording a vector overflow, before failing to match
2041    "c" and backing up to try the second alternative. The zero return, however,
2042    does correctly indicate that the maximum number of slots (namely 2) have been
2043    filled. In similar cases where there is temporary overflow, but the final
2044    number of used slots is actually less than the maximum, a non-zero value is
2045    returned.
2046  .P  .P
2047  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
2048  subpatterns there are in a compiled pattern. The smallest size for  subpatterns there are in a compiled pattern. The smallest size for
2049  \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
2050  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 1501  Offset values that correspond to unused Line 2060  Offset values that correspond to unused
2060  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
2061  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
2062  return from the function is 2, because the highest used capturing subpattern  return from the function is 2, because the highest used capturing subpattern
2063  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
2064  capturing subpatterns if you wish (assuming the vector is large enough, of  (assuming the vector is large enough, of course) are set to -1.
2065  course).  .P
2066    \fBNote\fP: Elements in the first two-thirds of \fIovector\fP that do not
2067    correspond to capturing parentheses in the pattern are never changed. That is,
2068    if a pattern contains \fIn\fP capturing parentheses, no more than
2069    \fIovector[0]\fP to \fIovector[2n+1]\fP are set by \fBpcre_exec()\fP. The other
2070    elements (in the first two-thirds) retain whatever values they previously had.
2071  .P  .P
2072  Some convenience functions are provided for extracting the captured substrings  Some convenience functions are provided for extracting the captured substrings
2073  as separate strings. These are described below.  as separate strings. These are described below.
2074  .  .
2075    .
2076  .\" HTML <a name="errorlist"></a>  .\" HTML <a name="errorlist"></a>
2077  .SS "Error return values from \fBpcre_exec()\fP"  .SS "Error return values from \fBpcre_exec()\fP"
2078  .rs  .rs
# Line 1549  If a pattern contains back references, b Line 2114  If a pattern contains back references, b
2114  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
2115  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
2116  automatically freed at the end of matching.  automatically freed at the end of matching.
2117    .P
2118    This error is also given if \fBpcre_stack_malloc()\fP fails in
2119    \fBpcre_exec()\fP. This can happen only when PCRE has been compiled with
2120    \fB--disable-stack-for-recursion\fP.
2121  .sp  .sp
2122    PCRE_ERROR_NOSUBSTRING    (-7)    PCRE_ERROR_NOSUBSTRING    (-7)
2123  .sp  .sp
# Line 1573  documentation for details. Line 2142  documentation for details.
2142  .sp  .sp
2143    PCRE_ERROR_BADUTF8        (-10)    PCRE_ERROR_BADUTF8        (-10)
2144  .sp  .sp
2145  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,
2146    and the PCRE_NO_UTF8_CHECK option was not set. If the size of the output vector
2147    (\fIovecsize\fP) is at least 2, the byte offset to the start of the the invalid
2148    UTF-8 character is placed in the first element, and a reason code is placed in
2149    the second element. The reason codes are listed in the
2150    .\" HTML <a href="#badutf8reasons">
2151    .\" </a>
2152    following section.
2153    .\"
2154    For backward compatibility, if PCRE_PARTIAL_HARD is set and the problem is a
2155    truncated UTF-8 character at the end of the subject (reason codes 1 to 5),
2156    PCRE_ERROR_SHORTUTF8 is returned instead of PCRE_ERROR_BADUTF8.
2157  .sp  .sp
2158    PCRE_ERROR_BADUTF8_OFFSET (-11)    PCRE_ERROR_BADUTF8_OFFSET (-11)
2159  .sp  .sp
2160  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
2161  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
2162    \fIstartoffset\fP did not point to the beginning of a UTF-8 character or the
2163    end of the subject.
2164  .sp  .sp
2165    PCRE_ERROR_PARTIAL        (-12)    PCRE_ERROR_PARTIAL        (-12)
2166  .sp  .sp
# Line 1592  documentation for details of partial mat Line 2174  documentation for details of partial mat
2174  .sp  .sp
2175  This code is no longer in use. It was formerly returned when the PCRE_PARTIAL  This code is no longer in use. It was formerly returned when the PCRE_PARTIAL
2176  option was used with a compiled pattern containing items that were not  option was used with a compiled pattern containing items that were not
2177  supported for partial matching. From release 8.00 onwards, there are no  supported for partial matching. From release 8.00 onwards, there are no
2178  restrictions on partial matching.  restrictions on partial matching.
2179  .sp  .sp
2180    PCRE_ERROR_INTERNAL       (-14)    PCRE_ERROR_INTERNAL       (-14)
# Line 1613  description above. Line 2195  description above.
2195    PCRE_ERROR_BADNEWLINE     (-23)    PCRE_ERROR_BADNEWLINE     (-23)
2196  .sp  .sp
2197  An invalid combination of PCRE_NEWLINE_\fIxxx\fP options was given.  An invalid combination of PCRE_NEWLINE_\fIxxx\fP options was given.
2198    .sp
2199      PCRE_ERROR_BADOFFSET      (-24)
2200    .sp
2201    The value of \fIstartoffset\fP was negative or greater than the length of the
2202    subject, that is, the value in \fIlength\fP.
2203    .sp
2204      PCRE_ERROR_SHORTUTF8      (-25)
2205    .sp
2206    This error is returned instead of PCRE_ERROR_BADUTF8 when the subject string
2207    ends with a truncated UTF-8 character and the PCRE_PARTIAL_HARD option is set.
2208    Information about the failure is returned as for PCRE_ERROR_BADUTF8. It is in
2209    fact sufficient to detect this case, but this special error code for
2210    PCRE_PARTIAL_HARD precedes the implementation of returned information; it is
2211    retained for backwards compatibility.
2212    .sp
2213      PCRE_ERROR_RECURSELOOP    (-26)
2214    .sp
2215    This error is returned when \fBpcre_exec()\fP detects a recursion loop within
2216    the pattern. Specifically, it means that either the whole pattern or a
2217    subpattern has been called recursively for the second time at the same position
2218    in the subject string. Some simple patterns that might do this are detected and
2219    faulted at compile time, but more complicated cases, in particular mutual
2220    recursions between two different subpatterns, cannot be detected until run
2221    time.
2222    .sp
2223      PCRE_ERROR_JIT_STACKLIMIT (-27)
2224    .sp
2225    This error is returned when a pattern that was successfully studied using a
2226    JIT compile option is being matched, but the memory available for the
2227    just-in-time processing stack is not large enough. See the
2228    .\" HREF
2229    \fBpcrejit\fP
2230    .\"
2231    documentation for more details.
2232    .sp
2233      PCRE_ERROR_BADMODE        (-28)
2234    .sp
2235    This error is given if a pattern that was compiled by the 8-bit library is
2236    passed to a 16-bit or 32-bit library function, or vice versa.
2237    .sp
2238      PCRE_ERROR_BADENDIANNESS  (-29)
2239    .sp
2240    This error is given if a pattern that was compiled and saved is reloaded on a
2241    host with different endianness. The utility function
2242    \fBpcre_pattern_to_host_byte_order()\fP can be used to convert such a pattern
2243    so that it runs on the new host.
2244    .P
2245    Error numbers -16 to -20, -22, and -30 are not used by \fBpcre_exec()\fP.
2246    .
2247    .
2248    .\" HTML <a name="badutf8reasons"></a>
2249    .SS "Reason codes for invalid UTF-8 strings"
2250    .rs
2251    .sp
2252    This section applies only to the 8-bit library. The corresponding information
2253    for the 16-bit library is given in the
2254    .\" HREF
2255    \fBpcre16\fP
2256    .\"
2257    page. The corresponding information for the 32-bit library is given in the
2258    .\" HREF
2259    \fBpcre32\fP
2260    .\"
2261    page.
2262  .P  .P
2263  Error numbers -16 to -20 and -22 are not used by \fBpcre_exec()\fP.  When \fBpcre_exec()\fP returns either PCRE_ERROR_BADUTF8 or
2264    PCRE_ERROR_SHORTUTF8, and the size of the output vector (\fIovecsize\fP) is at
2265    least 2, the offset of the start of the invalid UTF-8 character is placed in
2266    the first output vector element (\fIovector[0]\fP) and a reason code is placed
2267    in the second element (\fIovector[1]\fP). The reason codes are given names in
2268    the \fBpcre.h\fP header file:
2269    .sp
2270      PCRE_UTF8_ERR1
2271      PCRE_UTF8_ERR2
2272      PCRE_UTF8_ERR3
2273      PCRE_UTF8_ERR4
2274      PCRE_UTF8_ERR5
2275    .sp
2276    The string ends with a truncated UTF-8 character; the code specifies how many
2277    bytes are missing (1 to 5). Although RFC 3629 restricts UTF-8 characters to be
2278    no longer than 4 bytes, the encoding scheme (originally defined by RFC 2279)
2279    allows for up to 6 bytes, and this is checked first; hence the possibility of
2280    4 or 5 missing bytes.
2281    .sp
2282      PCRE_UTF8_ERR6
2283      PCRE_UTF8_ERR7
2284      PCRE_UTF8_ERR8
2285      PCRE_UTF8_ERR9
2286      PCRE_UTF8_ERR10
2287    .sp
2288    The two most significant bits of the 2nd, 3rd, 4th, 5th, or 6th byte of the
2289    character do not have the binary value 0b10 (that is, either the most
2290    significant bit is 0, or the next bit is 1).
2291    .sp
2292      PCRE_UTF8_ERR11
2293      PCRE_UTF8_ERR12
2294    .sp
2295    A character that is valid by the RFC 2279 rules is either 5 or 6 bytes long;
2296    these code points are excluded by RFC 3629.
2297    .sp
2298      PCRE_UTF8_ERR13
2299    .sp
2300    A 4-byte character has a value greater than 0x10fff; these code points are
2301    excluded by RFC 3629.
2302    .sp
2303      PCRE_UTF8_ERR14
2304    .sp
2305    A 3-byte character has a value in the range 0xd800 to 0xdfff; this range of
2306    code points are reserved by RFC 3629 for use with UTF-16, and so are excluded
2307    from UTF-8.
2308    .sp
2309      PCRE_UTF8_ERR15
2310      PCRE_UTF8_ERR16
2311      PCRE_UTF8_ERR17
2312      PCRE_UTF8_ERR18
2313      PCRE_UTF8_ERR19
2314    .sp
2315    A 2-, 3-, 4-, 5-, or 6-byte character is "overlong", that is, it codes for a
2316    value that can be represented by fewer bytes, which is invalid. For example,
2317    the two bytes 0xc0, 0xae give the value 0x2e, whose correct coding uses just
2318    one byte.
2319    .sp
2320      PCRE_UTF8_ERR20
2321    .sp
2322    The two most significant bits of the first byte of a character have the binary
2323    value 0b10 (that is, the most significant bit is 1 and the second is 0). Such a
2324    byte can only validly occur as the second or subsequent byte of a multi-byte
2325    character.
2326    .sp
2327      PCRE_UTF8_ERR21
2328    .sp
2329    The first byte of a character has the value 0xfe or 0xff. These values can
2330    never occur in a valid UTF-8 string.
2331  .  .
2332  .  .
2333  .SH "EXTRACTING CAPTURED SUBSTRINGS BY NUMBER"  .SH "EXTRACTING CAPTURED SUBSTRINGS BY NUMBER"
# Line 1764  then call \fBpcre_copy_substring()\fP or Line 2477  then call \fBpcre_copy_substring()\fP or
2477  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,
2478  the behaviour may not be what you want (see the next section).  the behaviour may not be what you want (see the next section).
2479  .P  .P
2480  \fBWarning:\fP If the pattern uses the "(?|" feature to set up multiple  \fBWarning:\fP If the pattern uses the (?| feature to set up multiple
2481  subpatterns with the same number, you cannot use names to distinguish them,  subpatterns with the same number, as described in the
2482  because names are not included in the compiled code. The matching process uses  .\" HTML <a href="pcrepattern.html#dupsubpatternnumber">
2483  only numbers.  .\" </a>
2484    section on duplicate subpattern numbers
2485    .\"
2486    in the
2487    .\" HREF
2488    \fBpcrepattern\fP
2489    .\"
2490    page, you cannot use names to distinguish the different subpatterns, because
2491    names are not included in the compiled code. The matching process uses only
2492    numbers. For this reason, the use of different names for subpatterns of the
2493    same number causes an error at compile time.
2494    .
2495  .  .
2496  .SH "DUPLICATE SUBPATTERN NAMES"  .SH "DUPLICATE SUBPATTERN NAMES"
2497  .rs  .rs
# Line 1777  only numbers. Line 2501  only numbers.
2501  .B const char *\fIname\fP, char **\fIfirst\fP, char **\fIlast\fP);  .B const char *\fIname\fP, char **\fIfirst\fP, char **\fIlast\fP);
2502  .PP  .PP
2503  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
2504  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
2505  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
2506  example is shown in the  such subpatterns are named, they are required to use the same names.)
2507    .P
2508    Normally, patterns with duplicate names are such that in any one match, only
2509    one of the named subpatterns participates. An example is shown in the
2510  .\" HREF  .\" HREF
2511  \fBpcrepattern\fP  \fBpcrepattern\fP
2512  .\"  .\"
# Line 1799  fourth are pointers to variables which a Line 2526  fourth are pointers to variables which a
2526  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
2527  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
2528  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
2529  described above in the section entitled \fIInformation about a pattern\fP.  described above in the section entitled \fIInformation about a pattern\fP
2530    .\" HTML <a href="#infoaboutpattern">
2531    .\" </a>
2532    above.
2533    .\"
2534  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
2535  numbers, and hence the captured data, if any.  numbers, and hence the captured data, if any.
2536  .  .
# Line 1826  other alternatives. Ultimately, when it Line 2557  other alternatives. Ultimately, when it
2557  will yield PCRE_ERROR_NOMATCH.  will yield PCRE_ERROR_NOMATCH.
2558  .  .
2559  .  .
2560    .SH "OBTAINING AN ESTIMATE OF STACK USAGE"
2561    .rs
2562    .sp
2563    Matching certain patterns using \fBpcre_exec()\fP can use a lot of process
2564    stack, which in certain environments can be rather limited in size. Some users
2565    find it helpful to have an estimate of the amount of stack that is used by
2566    \fBpcre_exec()\fP, to help them set recursion limits, as described in the
2567    .\" HREF
2568    \fBpcrestack\fP
2569    .\"
2570    documentation. The estimate that is output by \fBpcretest\fP when called with
2571    the \fB-m\fP and \fB-C\fP options is obtained by calling \fBpcre_exec\fP with
2572    the values NULL, NULL, NULL, -999, and -999 for its first five arguments.
2573    .P
2574    Normally, if its first argument is NULL, \fBpcre_exec()\fP immediately returns
2575    the negative error code PCRE_ERROR_NULL, but with this special combination of
2576    arguments, it returns instead a negative number whose absolute value is the
2577    approximate stack frame size in bytes. (A negative number is used so that it is
2578    clear that no match has happened.) The value is approximate because in some
2579    cases, recursive calls to \fBpcre_exec()\fP occur when there are one or two
2580    additional variables on the stack.
2581    .P
2582    If PCRE has been compiled to use the heap instead of the stack for recursion,
2583    the value returned is the size of each block that is obtained from the heap.
2584    .
2585    .
2586  .\" HTML <a name="dfamatch"></a>  .\" HTML <a name="dfamatch"></a>
2587  .SH "MATCHING A PATTERN: THE ALTERNATIVE FUNCTION"  .SH "MATCHING A PATTERN: THE ALTERNATIVE FUNCTION"
2588  .rs  .rs
# Line 1843  a compiled pattern, using a matching alg Line 2600  a compiled pattern, using a matching alg
2600  just once, and does not backtrack. This has different characteristics to the  just once, and does not backtrack. This has different characteristics to the
2601  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
2602  patterns are not supported. Nevertheless, there are times when this kind of  patterns are not supported. Nevertheless, there are times when this kind of
2603  matching can be useful. For a discussion of the two matching algorithms, and a  matching can be useful. For a discussion of the two matching algorithms, and a
2604  list of features that \fBpcre_dfa_exec()\fP does not support, see the  list of features that \fBpcre_dfa_exec()\fP does not support, see the
2605  .\" HREF  .\" HREF
2606  \fBpcrematching\fP  \fBpcrematching\fP
# Line 1884  Here is an example of a simple call to \ Line 2641  Here is an example of a simple call to \
2641  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
2642  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,
2643  PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NOTEMPTY_ATSTART,  PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NOTEMPTY_ATSTART,
2644  PCRE_NO_UTF8_CHECK, PCRE_PARTIAL_HARD, PCRE_PARTIAL_SOFT, PCRE_DFA_SHORTEST,  PCRE_NO_UTF8_CHECK, PCRE_BSR_ANYCRLF, PCRE_BSR_UNICODE, PCRE_NO_START_OPTIMIZE,
2645  and PCRE_DFA_RESTART. All but the last four of these are exactly the same as  PCRE_PARTIAL_HARD, PCRE_PARTIAL_SOFT, PCRE_DFA_SHORTEST, and PCRE_DFA_RESTART.
2646  for \fBpcre_exec()\fP, so their description is not repeated here.  All but the last four of these are exactly the same as for \fBpcre_exec()\fP,
2647    so their description is not repeated here.
2648  .sp  .sp
2649    PCRE_PARTIAL_HARD    PCRE_PARTIAL_HARD
2650    PCRE_PARTIAL_SOFT    PCRE_PARTIAL_SOFT
2651  .sp  .sp
2652  These have the same general effect as they do for \fBpcre_exec()\fP, but the  These have the same general effect as they do for \fBpcre_exec()\fP, but the
2653  details are slightly different. When PCRE_PARTIAL_HARD is set for  details are slightly different. When PCRE_PARTIAL_HARD is set for
# Line 1901  is converted into PCRE_ERROR_PARTIAL if Line 2659  is converted into PCRE_ERROR_PARTIAL if
2659  there have been no complete matches, but there is still at least one matching  there have been no complete matches, but there is still at least one matching
2660  possibility. The portion of the string that was inspected when the longest  possibility. The portion of the string that was inspected when the longest
2661  partial match was found is set as the first matching string in both cases.  partial match was found is set as the first matching string in both cases.
2662    There is a more detailed discussion of partial and multi-segment matching, with
2663    examples, in the
2664    .\" HREF
2665    \fBpcrepartial\fP
2666    .\"
2667    documentation.
2668  .sp  .sp
2669    PCRE_DFA_SHORTEST    PCRE_DFA_SHORTEST
2670  .sp  .sp
# Line 1922  match. There is more discussion of this Line 2686  match. There is more discussion of this
2686  .\"  .\"
2687  documentation.  documentation.
2688  .  .
2689    .
2690  .SS "Successful returns from \fBpcre_dfa_exec()\fP"  .SS "Successful returns from \fBpcre_dfa_exec()\fP"
2691  .rs  .rs
2692  .sp  .sp
# Line 1953  returns data, even though the meaning of Line 2718  returns data, even though the meaning of
2718  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
2719  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
2720  \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
2721  the longest matches.  the longest matches. Unlike \fBpcre_exec()\fP, \fBpcre_dfa_exec()\fP can use
2722    the entire \fIovector\fP for returning matched strings.
2723    .
2724  .  .
2725  .SS "Error returns from \fBpcre_dfa_exec()\fP"  .SS "Error returns from \fBpcre_dfa_exec()\fP"
2726  .rs  .rs
# Line 1982  group. These are not supported. Line 2749  group. These are not supported.
2749    PCRE_ERROR_DFA_UMLIMIT    (-18)    PCRE_ERROR_DFA_UMLIMIT    (-18)
2750  .sp  .sp
2751  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
2752  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
2753  supported (it is meaningless).  \fImatch_limit_recursion\fP fields. This is not supported (these fields are
2754    meaningless for DFA matching).
2755  .sp  .sp
2756    PCRE_ERROR_DFA_WSSIZE     (-19)    PCRE_ERROR_DFA_WSSIZE     (-19)
2757  .sp  .sp
# Line 1996  When a recursive subpattern is processed Line 2764  When a recursive subpattern is processed
2764  recursively, using private vectors for \fIovector\fP and \fIworkspace\fP. This  recursively, using private vectors for \fIovector\fP and \fIworkspace\fP. This
2765  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
2766  extremely rare, as a vector of size 1000 is used.  extremely rare, as a vector of size 1000 is used.
2767    .sp
2768      PCRE_ERROR_DFA_BADRESTART (-30)
2769    .sp
2770    When \fBpcre_dfa_exec()\fP is called with the \fBPCRE_DFA_RESTART\fP option,
2771    some plausibility checks are made on the contents of the workspace, which
2772    should contain data about the previous partial match. If any of these checks
2773    fail, this error is given.
2774  .  .
2775  .  .
2776  .SH "SEE ALSO"  .SH "SEE ALSO"
2777  .rs  .rs
2778  .sp  .sp
2779  \fBpcrebuild\fP(3), \fBpcrecallout\fP(3), \fBpcrecpp(3)\fP(3),  \fBpcre16\fP(3), \fBpcre32\fP(3), \fBpcrebuild\fP(3), \fBpcrecallout\fP(3),
2780  \fBpcrematching\fP(3), \fBpcrepartial\fP(3), \fBpcreposix\fP(3),  \fBpcrecpp(3)\fP(3), \fBpcrematching\fP(3), \fBpcrepartial\fP(3),
2781  \fBpcreprecompile\fP(3), \fBpcresample\fP(3), \fBpcrestack\fP(3).  \fBpcreposix\fP(3), \fBpcreprecompile\fP(3), \fBpcresample\fP(3),
2782    \fBpcrestack\fP(3).
2783  .  .
2784  .  .
2785  .SH AUTHOR  .SH AUTHOR
# Line 2020  Cambridge CB2 3QH, England. Line 2796  Cambridge CB2 3QH, England.
2796  .rs  .rs
2797  .sp  .sp
2798  .nf  .nf
2799  Last updated: 11 September 2009  Last updated: 07 September 2012
2800  Copyright (c) 1997-2009 University of Cambridge.  Copyright (c) 1997-2012 University of Cambridge.
2801  .fi  .fi

Legend:
Removed from v.442  
changed lines
  Added in v.1080

  ViewVC Help
Powered by ViewVC 1.1.5