/[pcre]/code/trunk/doc/pcreapi.3

Diff of /code/trunk/doc/pcreapi.3

Parent Directory | Revision Log | View Patch Patch

-revision 75 by nigel,
Sat Feb 24 21:40:37 2007 UTC
+revision 686 by ph10,
Tue Sep  6 16:06:36 2011 UTC
 Line 1
- .TH PCRE 3
+ .TH PCREAPI 3
  .SH NAME
  PCRE - Perl-compatible regular expressions
- .SH "PCRE NATIVE API"
+ .SH "PCRE NATIVE API BASIC FUNCTIONS"
  .rs
  .sp
  .B #include <pcre.h>
  .PP
  .SM
- .br
  .B pcre *pcre_compile(const char *\fIpattern\fP, int \fIoptions\fP,
  .ti +5n
  .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
  .ti +5n
  .B const unsigned char *\fItableptr\fP);
  .PP
- .br
+ .B pcre *pcre_compile2(const char *\fIpattern\fP, int \fIoptions\fP,
+ .ti +5n
+ .B int *\fIerrorcodeptr\fP,
+ .ti +5n
+ .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
+ .ti +5n
+ .B const unsigned char *\fItableptr\fP);
+ .PP
  .B pcre_extra *pcre_study(const pcre *\fIcode\fP, int \fIoptions\fP,
  .ti +5n
  .B const char **\fIerrptr\fP);
  .PP
- .br
+ .B void pcre_free_study(pcre_extra *\fIextra\fP);
+ .PP
  .B int pcre_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
  .ti +5n
  .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
  .ti +5n
  .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP);
+ .
+ .
+ .SH "PCRE NATIVE API AUXILIARY FUNCTIONS"
+ .rs
+ .sp
+ .B pcre_jit_stack *pcre_jit_stack_alloc(int \fIstartsize\fP, int \fImaxsize\fP);
+ .PP
+ .B void pcre_jit_stack_free(pcre_jit_stack *\fIstack\fP);
+ .PP
+ .B void pcre_assign_jit_stack(pcre_extra *\fIextra\fP,
+ .ti +5n
+ .B pcre_jit_callback \fIcallback\fP, void *\fIdata\fP);
+ .PP
+ .B int pcre_dfa_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
+ .ti +5n
+ .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
+ .ti +5n
+ .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP,
+ .ti +5n
+ .B int *\fIworkspace\fP, int \fIwscount\fP);
  .PP
- .br
  .B int pcre_copy_named_substring(const pcre *\fIcode\fP,
  .ti +5n
  .B const char *\fIsubject\fP, int *\fIovector\fP,
-Line 35 
 PCRE - Perl-compatible regular expressio
+Line 61 
 PCRE - Perl-compatible regular expressio
  .ti +5n
  .B char *\fIbuffer\fP, int \fIbuffersize\fP);
  .PP
- .br
  .B int pcre_copy_substring(const char *\fIsubject\fP, int *\fIovector\fP,
  .ti +5n
  .B int \fIstringcount\fP, int \fIstringnumber\fP, char *\fIbuffer\fP,
  .ti +5n
  .B int \fIbuffersize\fP);
  .PP
- .br
  .B int pcre_get_named_substring(const pcre *\fIcode\fP,
  .ti +5n
  .B const char *\fIsubject\fP, int *\fIovector\fP,
-Line 51 
 PCRE - Perl-compatible regular expressio
+Line 75 
 PCRE - Perl-compatible regular expressio
  .ti +5n
  .B const char **\fIstringptr\fP);
  .PP
- .br
  .B int pcre_get_stringnumber(const pcre *\fIcode\fP,
  .ti +5n
  .B const char *\fIname\fP);
  .PP
- .br
+ .B int pcre_get_stringtable_entries(const pcre *\fIcode\fP,
+ .ti +5n
+ .B const char *\fIname\fP, char **\fIfirst\fP, char **\fIlast\fP);
+ .PP
  .B int pcre_get_substring(const char *\fIsubject\fP, int *\fIovector\fP,
  .ti +5n
  .B int \fIstringcount\fP, int \fIstringnumber\fP,
  .ti +5n
  .B const char **\fIstringptr\fP);
  .PP
- .br
  .B int pcre_get_substring_list(const char *\fIsubject\fP,
  .ti +5n
  .B int *\fIovector\fP, int \fIstringcount\fP, "const char ***\fIlistptr\fP);"
  .PP
- .br
  .B void pcre_free_substring(const char *\fIstringptr\fP);
  .PP
- .br
  .B void pcre_free_substring_list(const char **\fIstringptr\fP);
  .PP
- .br
  .B const unsigned char *pcre_maketables(void);
  .PP
- .br
  .B int pcre_fullinfo(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
  .ti +5n
  .B int \fIwhat\fP, void *\fIwhere\fP);
  .PP
- .br
  .B int pcre_info(const pcre *\fIcode\fP, int *\fIoptptr\fP, int
  .B *\fIfirstcharptr\fP);
  .PP
- .br
+ .B int pcre_refcount(pcre *\fIcode\fP, int \fIadjust\fP);
+ .PP
  .B int pcre_config(int \fIwhat\fP, void *\fIwhere\fP);
  .PP
- .br
  .B char *pcre_version(void);
- .PP
+ .
- .br
+ .
+ .SH "PCRE NATIVE API INDIRECTED FUNCTIONS"
+ .rs
+ .sp
  .B void *(*pcre_malloc)(size_t);
  .PP
- .br
  .B void (*pcre_free)(void *);
  .PP
- .br
  .B void *(*pcre_stack_malloc)(size_t);
  .PP
- .br
  .B void (*pcre_stack_free)(void *);
  .PP
- .br
  .B int (*pcre_callout)(pcre_callout_block *);
  .
  .
  .SH "PCRE API OVERVIEW"
  .rs
  .sp
- PCRE has its own native API, which is described in this document. There is also
+ PCRE has its own native API, which is described in this document. There are
- a set of wrapper functions that correspond to the POSIX regular expression API.
+ also some wrapper functions that correspond to the POSIX regular expression
- These are described in the
+ API, but they do not give access to all the functionality. They are described
+ in the
  .\" HREF
  \fBpcreposix\fP
  .\"
- documentation.
+ documentation. Both of these APIs define a set of C function calls. A C++
+ wrapper is also distributed with PCRE. It is documented in the
+ .\" HREF
+ \fBpcrecpp\fP
+ .\"
+ page.
  .P
- The native API function prototypes are defined in the header file \fBpcre.h\fP,
+ The native API C function prototypes are defined in the header file
- and on Unix systems the library itself is called \fBlibpcre\fP. It can
+ \fBpcre.h\fP, and on Unix systems the library itself is called \fBlibpcre\fP.
- normally be accessed by adding \fB-lpcre\fP to the command for linking an
+ It can normally be accessed by adding \fB-lpcre\fP to the command for linking
- application that uses PCRE. The header file defines the macros PCRE_MAJOR and
+ an application that uses PCRE. The header file defines the macros PCRE_MAJOR
- PCRE_MINOR to contain the major and minor release numbers for the library.
+ and PCRE_MINOR to contain the major and minor release numbers for the library.
  Applications can use these to include support for different releases of PCRE.
  .P
- The functions \fBpcre_compile()\fP, \fBpcre_study()\fP, and \fBpcre_exec()\fP
+ In a Windows environment, if you want to statically link an application program
- are used for compiling and matching regular expressions. A sample program that
+ against a non-dll \fBpcre.a\fP file, you must define PCRE_STATIC before
- demonstrates the simplest way of using them is provided in the file called
+ including \fBpcre.h\fP or \fBpcrecpp.h\fP, because otherwise the
- \fIpcredemo.c\fP in the source distribution. The
+ \fBpcre_malloc()\fP and \fBpcre_free()\fP exported functions will be declared
+ \fB__declspec(dllimport)\fP, with unwanted results.
+ .P
+ The functions \fBpcre_compile()\fP, \fBpcre_compile2()\fP, \fBpcre_study()\fP,
+ and \fBpcre_exec()\fP are used for compiling and matching regular expressions
+ in a Perl-compatible manner. A sample program that demonstrates the simplest
+ way of using them is provided in the file called \fIpcredemo.c\fP in the PCRE
+ source distribution. A listing of this program is given in the
+ .\" HREF
+ \fBpcredemo\fP
+ .\"
+ documentation, and the
  .\" HREF
  \fBpcresample\fP
  .\"
- documentation describes how to run it.
+ documentation describes how to compile and run it.
+ .P
+ Just-in-time compiler support is an optional feature of PCRE that can be built
+ in appropriate hardware environments. It greatly speeds up the matching
+ performance of many patterns. Simple programs can easily request that it be
+ used if available, by setting an option that is ignored when it is not
+ relevant. More complicated programs might need to make use of the functions
+ \fBpcre_jit_stack_alloc()\fP, \fBpcre_jit_stack_free()\fP, and
+ \fBpcre_assign_jit_stack()\fP in order to control the JIT code's memory usage.
+ These functions are discussed in the
+ .\" HREF
+ \fBpcrejit\fP
+ .\"
+ documentation.
+ .P
+ A second matching function, \fBpcre_dfa_exec()\fP, which is not
+ Perl-compatible, is also provided. This uses a different algorithm for the
+ matching. The alternative algorithm finds all possible matches (at a given
+ point in the subject), and scans the subject just once (unless there are
+ lookbehind assertions). However, this algorithm does not return captured
+ substrings. A description of the two matching algorithms and their advantages
+ and disadvantages is given in the
+ .\" HREF
+ \fBpcrematching\fP
+ .\"
+ documentation.
  .P
  In addition to the main compiling and matching functions, there are convenience
- functions for extracting captured substrings from a matched subject string.
+ functions for extracting captured substrings from a subject string that is
- They are:
+ matched by \fBpcre_exec()\fP. They are:
  .sp
    \fBpcre_copy_substring()\fP
    \fBpcre_copy_named_substring()\fP
-Line 145 
 They are:
+Line 206 
 They are:
    \fBpcre_get_named_substring()\fP
    \fBpcre_get_substring_list()\fP
    \fBpcre_get_stringnumber()\fP
+   \fBpcre_get_stringtable_entries()\fP
  .sp
  \fBpcre_free_substring()\fP and \fBpcre_free_substring_list()\fP are also
  provided, to free the memory used for extracted strings.
  .P
  The function \fBpcre_maketables()\fP is used to build a set of character tables
- in the current locale for passing to \fBpcre_compile()\fP or \fBpcre_exec()\fP.
+ in the current locale for passing to \fBpcre_compile()\fP, \fBpcre_exec()\fP,
- This is an optional facility that is provided for specialist use. Most
+ or \fBpcre_dfa_exec()\fP. This is an optional facility that is provided for
- commonly, no special tables are passed, in which case internal tables that are
+ specialist use. Most commonly, no special tables are passed, in which case
- generated when PCRE is built are used.
+ internal tables that are generated when PCRE is built are used.
  .P
  The function \fBpcre_fullinfo()\fP is used to find out information about a
  compiled pattern; \fBpcre_info()\fP is an obsolete version that returns only
-Line 161 
 some of the available information, but i
+Line 223 
 some of the available information, but i
  The function \fBpcre_version()\fP returns a pointer to a string containing the
  version of PCRE and its date of release.
  .P
+ The function \fBpcre_refcount()\fP maintains a reference count in a data block
+ containing a compiled pattern. This is provided for the benefit of
+ object-oriented applications.
+ .P
  The global variables \fBpcre_malloc\fP and \fBpcre_free\fP initially contain
  the entry points of the standard \fBmalloc()\fP and \fBfree()\fP functions,
  respectively. PCRE calls the memory management functions via these variables,
-Line 170 
 should be done before calling any PCRE f
+Line 236 
 should be done before calling any PCRE f
  The global variables \fBpcre_stack_malloc\fP and \fBpcre_stack_free\fP are also
  indirections to memory management functions. These special functions are used
  only when PCRE is compiled to use the heap for remembering data, instead of
- recursive function calls. This is a non-standard way of building PCRE, for use
+ recursive function calls, when running the \fBpcre_exec()\fP function. See the
- in environments that have limited stacks. Because of the greater use of memory
+ .\" HREF
- management, it runs more slowly. Separate functions are provided so that
+ \fBpcrebuild\fP
- special-purpose external code can be used for this case. When used, these
+ .\"
- functions are always called in a stack-like manner (last obtained, first
+ documentation for details of how to do this. It is a non-standard way of
- freed), and always for memory blocks of the same size.
+ building PCRE, for use in environments that have limited stacks. Because of the
+ greater use of memory management, it runs more slowly. Separate functions are
+ provided so that special-purpose external code can be used for this case. When
+ used, these functions are always called in a stack-like manner (last obtained,
+ first freed), and always for memory blocks of the same size. There is a
+ discussion about PCRE's stack usage in the
+ .\" HREF
+ \fBpcrestack\fP
+ .\"
+ documentation.
  .P
  The global variable \fBpcre_callout\fP initially contains NULL. It can be set
  by the caller to a "callout" function, which PCRE will then call at specified
-Line 186 
 points during a matching operation. Deta
+Line 261 
 points during a matching operation. Deta
  documentation.
  .
  .
+ .\" HTML <a name="newlines"></a>
+ .SH NEWLINES
+ .rs
+ .sp
+ PCRE supports five different conventions for indicating line breaks in
+ strings: a single CR (carriage return) character, a single LF (linefeed)
+ character, the two-character sequence CRLF, any of the three preceding, or any
+ Unicode newline sequence. The Unicode newline sequences are the three just
+ mentioned, plus the single characters VT (vertical tab, U+000B), FF (formfeed,
+ U+000C), NEL (next line, U+0085), LS (line separator, U+2028), and PS
+ (paragraph separator, U+2029).
+ .P
+ Each of the first three conventions is used by at least one operating system as
+ its standard newline sequence. When PCRE is built, a default can be specified.
+ The default default is LF, which is the Unix standard. When PCRE is run, the
+ default can be overridden, either when a pattern is compiled, or when it is
+ matched.
+ .P
+ At compile time, the newline convention can be specified by the \fIoptions\fP
+ argument of \fBpcre_compile()\fP, or it can be specified by special text at the
+ start of the pattern itself; this overrides any other settings. See the
+ .\" HREF
+ \fBpcrepattern\fP
+ .\"
+ page for details of the special character sequences.
+ .P
+ In the PCRE documentation the word "newline" is used to mean "the character or
+ pair of characters that indicate a line break". The choice of newline
+ convention affects the handling of the dot, circumflex, and dollar
+ metacharacters, the handling of #-comments in /x mode, and, when CRLF is a
+ recognized line ending sequence, the match position advancement for a
+ non-anchored pattern. There is more detail about this in the
+ .\" HTML <a href="#execoptions">
+ .\" </a>
+ section on \fBpcre_exec()\fP options
+ .\"
+ below.
+ .P
+ The choice of newline convention does not affect the interpretation of
+ the \en or \er escape sequences, nor does it affect what \eR matches, which is
+ controlled in a similar way, but by separate options.
+ .
+ .
  .SH MULTITHREADING
  .rs
  .sp
-Line 196 
 callout function pointed to by \fBpcre_c
+Line 314 
 callout function pointed to by \fBpcre_c
  .P
  The compiled form of a regular expression is not altered during matching, so
  the same compiled pattern can safely be used by several threads at once.
+ .P
+ If the just-in-time optimization feature is being used, it needs separate
+ memory stack areas for each thread. See the
+ .\" HREF
+ \fBpcrejit\fP
+ .\"
+ documentation for more details.
  .
  .
  .SH "SAVING PRECOMPILED PATTERNS FOR LATER USE"
-Line 207 
 which it was compiled. Details are given
+Line 332 
 which it was compiled. Details are given
  .\" HREF
  \fBpcreprecompile\fP
  .\"
- documentation.
+ documentation. However, compiling a regular expression with one version of PCRE
+ for use with a different version is not guaranteed to work and may cause
+ crashes.
  .
  .
  .SH "CHECKING BUILD-TIME OPTIONS"
-Line 236 
 otherwise it is set to zero.
+Line 363 
 otherwise it is set to zero.
  The output is an integer that is set to one if support for Unicode character
  properties is available; otherwise it is set to zero.
  .sp
+   PCRE_CONFIG_JIT
+ .sp
+ The output is an integer that is set to one if support for just-in-time
+ compiling is available; otherwise it is set to zero.
+ .sp
    PCRE_CONFIG_NEWLINE
  .sp
- The output is an integer that is set to the value of the code that is used for
+ The output is an integer whose value specifies the default character sequence
- the newline character. It is either linefeed (10) or carriage return (13), and
+ that is recognized as meaning "newline". The four values that are supported
- should normally be the standard character for your operating system.
+ are: 10 for LF, 13 for CR, 3338 for CRLF, -2 for ANYCRLF, and -1 for ANY.
+ Though they are derived from ASCII, the same values are returned in EBCDIC
+ environments. The default should normally correspond to the standard sequence
+ for your operating system.
+ .sp
+   PCRE_CONFIG_BSR
+ .sp
+ The output is an integer whose value indicates what character sequences the \eR
+ escape sequence matches by default. A value of 0 means that \eR matches any
+ Unicode line ending sequence; a value of 1 means that \eR matches only CR, LF,
+ or CRLF. The default can be overridden when a pattern is compiled or matched.
  .sp
    PCRE_CONFIG_LINK_SIZE
  .sp
-Line 262 
 documentation.
+Line 404 
 documentation.
  .sp
    PCRE_CONFIG_MATCH_LIMIT
  .sp
- The output is an integer that gives the default limit for the number of
+ The output is a long integer that gives the default limit for the number of
  internal matching function calls in a \fBpcre_exec()\fP execution. Further
  details are given with \fBpcre_exec()\fP below.
  .sp
+   PCRE_CONFIG_MATCH_LIMIT_RECURSION
+ .sp
+ The output is a long integer that gives the default limit for the depth of
+ recursion when calling the internal matching function in a \fBpcre_exec()\fP
+ execution. Further details are given with \fBpcre_exec()\fP below.
+ .sp
    PCRE_CONFIG_STACKRECURSE
  .sp
- The output is an integer that is set to one if internal recursion is
+ The output is an integer that is set to one if internal recursion when running
- implemented by recursive function calls that use the stack to remember their
+ \fBpcre_exec()\fP is implemented by recursive function calls that use the stack
- state. This is the usual way that PCRE is compiled. The output is zero if PCRE
+ to remember their state. This is the usual way that PCRE is compiled. The
- was compiled to use blocks of data on the heap instead of recursive function
+ output is zero if PCRE was compiled to use blocks of data on the heap instead
- calls. In this case, \fBpcre_stack_malloc\fP and \fBpcre_stack_free\fP are
+ of recursive function calls. In this case, \fBpcre_stack_malloc\fP and
- called to manage memory blocks on the heap, thus avoiding the use of the stack.
+ \fBpcre_stack_free\fP are called to manage memory blocks on the heap, thus
+ avoiding the use of the stack.
  .
  .
  .SH "COMPILING A PATTERN"
-Line 284 
 called to manage memory blocks on the he
+Line 433 
 called to manage memory blocks on the he
  .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
  .ti +5n
  .B const unsigned char *\fItableptr\fP);
+ .sp
+ .B pcre *pcre_compile2(const char *\fIpattern\fP, int \fIoptions\fP,
+ .ti +5n
+ .B int *\fIerrorcodeptr\fP,
+ .ti +5n
+ .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
+ .ti +5n
+ .B const unsigned char *\fItableptr\fP);
  .P
- The function \fBpcre_compile()\fP is called to compile a pattern into an
+ Either of the functions \fBpcre_compile()\fP or \fBpcre_compile2()\fP can be
- internal form. The pattern is a C string terminated by a binary zero, and
+ called to compile a pattern into an internal form. The only difference between
- is passed in the \fIpattern\fP argument. A pointer to a single block of memory
+ the two interfaces is that \fBpcre_compile2()\fP has an additional argument,
- that is obtained via \fBpcre_malloc\fP is returned. This contains the compiled
+ \fIerrorcodeptr\fP, via which a numerical error code can be returned. To avoid
- code and related data. The \fBpcre\fP type is defined for the returned block;
+ too much repetition, we refer just to \fBpcre_compile()\fP below, but the
- this is a typedef for a structure whose contents are not externally defined. It
+ information applies equally to \fBpcre_compile2()\fP.
- is up to the caller to free the memory when it is no longer required.
+ .P
+ The pattern is a C string terminated by a binary zero, and is passed in the
+ \fIpattern\fP argument. A pointer to a single block of memory that is obtained
+ via \fBpcre_malloc\fP is returned. This contains the compiled code and related
+ data. The \fBpcre\fP type is defined for the returned block; this is a typedef
+ for a structure whose contents are not externally defined. It is up to the
+ caller to free the memory (via \fBpcre_free\fP) when it is no longer required.
  .P
  Although the compiled code of a PCRE regex is relocatable, that is, it does not
  depend on memory location, the complete \fBpcre\fP data block is not
  fully relocatable, because it may contain a copy of the \fItableptr\fP
  argument, which is an address (see below).
  .P
- The \fIoptions\fP argument contains independent bits that affect the
+ The \fIoptions\fP argument contains various bit settings that affect the
  compilation. It should be zero if no options are required. The available
- options are described below. Some of them, in particular, those that are
+ options are described below. Some of them (in particular, those that are
- compatible with Perl, can also be set and unset from within the pattern (see
+ compatible with Perl, but some others as well) can also be set and unset from
- the detailed description in the
+ within the pattern (see the detailed description in the
  .\" HREF
  \fBpcrepattern\fP
  .\"
- documentation). For these options, the contents of the \fIoptions\fP argument
+ documentation). For those options that can be different in different parts of
- specifies their initial settings at the start of compilation and execution. The
+ the pattern, the contents of the \fIoptions\fP argument specifies their
- PCRE_ANCHORED option can be set at the time of matching as well as at compile
+ settings at the start of compilation and execution. The PCRE_ANCHORED,
- time.
+ PCRE_BSR_\fIxxx\fP, PCRE_NEWLINE_\fIxxx\fP, PCRE_NO_UTF8_CHECK, and
+ PCRE_NO_START_OPT options can be set at the time of matching as well as at
+ compile time.
  .P
  If \fIerrptr\fP is NULL, \fBpcre_compile()\fP returns NULL immediately.
  Otherwise, if compilation of a pattern fails, \fBpcre_compile()\fP returns
  NULL, and sets the variable pointed to by \fIerrptr\fP to point to a textual
- error message. The offset from the start of the pattern to the character where
+ error message. This is a static string that is part of the library. You must
- the error was discovered is placed in the variable pointed to by
+ not try to free it. Normally, the offset from the start of the pattern to the
- \fIerroffset\fP, which must not be NULL. If it is, an immediate error is given.
+ byte that was being processed when the error was discovered is placed in the
+ variable pointed to by \fIerroffset\fP, which must not be NULL (if it is, an
+ immediate error is given). However, for an invalid UTF-8 string, the offset is
+ that of the first byte of the failing character. Also, some errors are not
+ detected until checks are carried out when the whole pattern has been scanned;
+ in these cases the offset passed back is the length of the pattern.
+ .P
+ Note that the offset is in bytes, not characters, even in UTF-8 mode. It may
+ sometimes point into the middle of a UTF-8 character.
+ .P
+ If \fBpcre_compile2()\fP is used instead of \fBpcre_compile()\fP, and the
+ \fIerrorcodeptr\fP argument is not NULL, a non-zero error code number is
+ returned via this argument in the event of an error. This is in addition to the
+ textual error message. Error codes and messages are listed below.
  .P
  If the final argument, \fItableptr\fP, is NULL, PCRE uses a default set of
  character tables that are built when PCRE is compiled, using the default C
-Line 358 
 facility, see the
+Line 536 
 facility, see the
  .\"
  documentation.
  .sp
+   PCRE_BSR_ANYCRLF
+   PCRE_BSR_UNICODE
+ .sp
+ These options (which are mutually exclusive) control what the \eR escape
+ sequence matches. The choice is either to match only CR, LF, or CRLF, or to
+ match any Unicode newline sequence. The default is specified when PCRE is
+ built. It can be overridden from within the pattern, or by setting an option
+ when a compiled pattern is matched.
+ .sp
    PCRE_CASELESS
  .sp
  If this bit is set, letters in the pattern match both upper and lower case
  letters. It is equivalent to Perl's /i option, and it can be changed within a
- pattern by a (?i) option setting. When running in UTF-8 mode, case support for
+ pattern by a (?i) option setting. In UTF-8 mode, PCRE always understands the
- high-valued characters is available only when PCRE is built with Unicode
+ concept of case for characters whose values are less than 128, so caseless
- character property support.
+ matching is always possible. For characters with higher values, the concept of
+ case is supported if PCRE is compiled with Unicode property support, but not
+ otherwise. If you want to use caseless matching for characters 128 and above,
+ you must ensure that PCRE is compiled with Unicode property support as well as
+ with UTF-8 support.
  .sp
    PCRE_DOLLAR_ENDONLY
  .sp
  If this bit is set, a dollar metacharacter in the pattern matches only at the
  end of the subject string. Without this option, a dollar also matches
- immediately before the final character if it is a newline (but not before any
+ immediately before a newline at the end of the string (but not before any other
- other newlines). The PCRE_DOLLAR_ENDONLY option is ignored if PCRE_MULTILINE is
+ newlines). The PCRE_DOLLAR_ENDONLY option is ignored if PCRE_MULTILINE is set.
- set. There is no equivalent to this option in Perl, and no way to set it within
+ There is no equivalent to this option in Perl, and no way to set it within a
- a pattern.
+ pattern.
  .sp
    PCRE_DOTALL
  .sp
- 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
- including newlines. Without it, newlines are excluded. This option is
+ any value, including one that indicates a newline. However, it only ever
+ matches one character, even if newlines are coded as CRLF. Without this option,
+ a dot does not match when the current position is at a newline. This option is
  equivalent to Perl's /s option, and it can be changed within a pattern by a
- (?s) option setting. A negative class such as [^a] always matches a newline
+ (?s) option setting. A negative class such as [^a] always matches newline
- character, independent of the setting of this option.
+ characters, independent of the setting of this option.
+ .sp
+   PCRE_DUPNAMES
+ .sp
+ If this bit is set, names used to identify capturing subpatterns need not be
+ unique. This can be helpful for certain types of pattern when it is known that
+ only one instance of the named subpattern can ever be matched. There are more
+ details of named subpatterns below; see also the
+ .\" HREF
+ \fBpcrepattern\fP
+ .\"
+ documentation.
  .sp
    PCRE_EXTENDED
  .sp
  If this bit is set, whitespace data characters in the pattern are totally
  ignored except when escaped or inside a character class. Whitespace does not
  include the VT character (code 11). In addition, characters between an
- unescaped # outside a character class and the next newline character,
+ unescaped # outside a character class and the next newline, inclusive, are also
- inclusive, are also ignored. This is equivalent to Perl's /x option, and it can
+ ignored. This is equivalent to Perl's /x option, and it can be changed within a
- be changed within a pattern by a (?x) option setting.
+ pattern by a (?x) option setting.
+ .P
+ Which characters are interpreted as newlines is controlled by the options
+ passed to \fBpcre_compile()\fP or by a special sequence at the start of the
+ pattern, as described in the section entitled
+ .\" HTML <a href="pcrepattern.html#newlines">
+ .\" </a>
+ "Newline conventions"
+ .\"
+ in the \fBpcrepattern\fP documentation. Note that the end of this type of
+ comment is a literal newline sequence in the pattern; escape sequences that
+ happen to represent a newline do not count.
  .P
  This option makes it possible to include comments inside complicated patterns.
  Note, however, that this applies only to data characters. Whitespace characters
  may never appear within special character sequences in a pattern, for example
- within the sequence (?( which introduces a conditional subpattern.
+ within the sequence (?( that introduces a conditional subpattern.
  .sp
    PCRE_EXTRA
  .sp
-Line 404 
 that is incompatible with Perl, but it i
+Line 619 
 that is incompatible with Perl, but it i
  set, any backslash in a pattern that is followed by a letter that has no
  special meaning causes an error, thus reserving these combinations for future
  expansion. By default, as in Perl, a backslash followed by a letter with no
- special meaning is treated as a literal. There are at present no other features
+ special meaning is treated as a literal. (Perl can, however, be persuaded to
- controlled by this option. It can also be set by a (?X) option setting within a
+ give an error for this, by running it with the -w option.) There are at present
- pattern.
+ no other features controlled by this option. It can also be set by a (?X)
+ option setting within a pattern.
+ .sp
+   PCRE_FIRSTLINE
+ .sp
+ If this option is set, an unanchored pattern is required to match before or at
+ the first newline in the subject string, though the matched text may continue
+ over the newline.
+ .sp
+   PCRE_JAVASCRIPT_COMPAT
+ .sp
+ If this option is set, PCRE's behaviour is changed in some ways so that it is
+ compatible with JavaScript rather than Perl. The changes are as follows:
+ .P
+ (1) A lone closing square bracket in a pattern causes a compile-time error,
+ because this is illegal in JavaScript (by default it is treated as a data
+ character). Thus, the pattern AB]CD becomes illegal when this option is set.
+ .P
+ (2) At run time, a back reference to an unset subpattern group matches an empty
+ string (by default this causes the current matching alternative to fail). A
+ pattern such as (\e1)(a) succeeds when this option is set (assuming it can find
+ an "a" in the subject), whereas it fails by default, for Perl compatibility.
  .sp
    PCRE_MULTILINE
  .sp
-Line 418 
 terminating newline (unless PCRE_DOLLAR_
+Line 654 
 terminating newline (unless PCRE_DOLLAR_
  Perl.
  .P
  When PCRE_MULTILINE it is set, the "start of line" and "end of line" constructs
- match immediately following or immediately before any newline in the subject
+ match immediately following or immediately before internal newlines in the
- string, respectively, as well as at the very start and end. This is equivalent
+ subject string, respectively, as well as at the very start and end. This is
- to Perl's /m option, and it can be changed within a pattern by a (?m) option
+ equivalent to Perl's /m option, and it can be changed within a pattern by a
- setting. If there are no "\en" characters in a subject string, or no
+ (?m) option setting. If there are no newlines in a subject string, or no
  occurrences of ^ or $ in a pattern, setting PCRE_MULTILINE has no effect.
  .sp
+   PCRE_NEWLINE_CR
+   PCRE_NEWLINE_LF
+   PCRE_NEWLINE_CRLF
+   PCRE_NEWLINE_ANYCRLF
+   PCRE_NEWLINE_ANY
+ .sp
+ These options override the default newline definition that was chosen when PCRE
+ was built. Setting the first or the second specifies that a newline is
+ indicated by a single character (CR or LF, respectively). Setting
+ PCRE_NEWLINE_CRLF specifies that a newline is indicated by the two-character
+ CRLF sequence. Setting PCRE_NEWLINE_ANYCRLF specifies that any of the three
+ preceding sequences should be recognized. Setting PCRE_NEWLINE_ANY specifies
+ that any Unicode newline sequence should be recognized. The Unicode newline
+ sequences are the three just mentioned, plus the single characters VT (vertical
+ tab, U+000B), FF (formfeed, U+000C), NEL (next line, U+0085), LS (line
+ separator, U+2028), and PS (paragraph separator, U+2029). The last two are
+ recognized only in UTF-8 mode.
+ .P
+ The newline setting in the options word uses three bits that are treated
+ as a number, giving eight possibilities. Currently only six are used (default
+ plus the five values above). This means that if you set more than one newline
+ option, the combination may or may not be sensible. For example,
+ PCRE_NEWLINE_CR with PCRE_NEWLINE_LF is equivalent to PCRE_NEWLINE_CRLF, but
+ other combinations may yield unused numbers and cause an error.
+ .P
+ The only time that a line break in a pattern is specially recognized when
+ compiling is when PCRE_EXTENDED is set. CR and LF are whitespace characters,
+ and so are ignored in this mode. Also, an unescaped # outside a character class
+ indicates a comment that lasts until after the next line break sequence. In
+ other circumstances, line break sequences in patterns are treated as literal
+ data.
+ .P
+ The newline option that is set at compile time becomes the default that is used
+ for \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP, but it can be overridden.
+ .sp
    PCRE_NO_AUTO_CAPTURE
  .sp
  If this option is set, it disables the use of numbered capturing parentheses in
-Line 432 
 were followed by ?: but named parenthese
+Line 703 
 were followed by ?: but named parenthese
  they acquire numbers in the usual way). There is no equivalent of this option
  in Perl.
  .sp
+   NO_START_OPTIMIZE
+ .sp
+ This is an option that acts at matching time; that is, it is really an option
+ for \fBpcre_exec()\fP or \fBpcre_dfa_exec()\fP. If it is set at compile time,
+ it is remembered with the compiled pattern and assumed at matching time. For
+ details see the discussion of PCRE_NO_START_OPTIMIZE
+ .\" HTML <a href="#execoptions">
+ .\" </a>
+ below.
+ .\"
+ .sp
+   PCRE_UCP
+ .sp
+ This option changes the way PCRE processes \eB, \eb, \eD, \ed, \eS, \es, \eW,
+ \ew, and some of the POSIX character classes. By default, only ASCII characters
+ are recognized, but if PCRE_UCP is set, Unicode properties are used instead to
+ classify characters. More details are given in the section on
+ .\" HTML <a href="pcre.html#genericchartypes">
+ .\" </a>
+ generic character types
+ .\"
+ in the
+ .\" HREF
+ \fBpcrepattern\fP
+ .\"
+ page. If you set PCRE_UCP, matching one of the items it affects takes much
+ longer. The option is available only if PCRE has been compiled with Unicode
+ property support.
+ .sp
    PCRE_UNGREEDY
  .sp
  This option inverts the "greediness" of the quantifiers so that they are not
-Line 445 
 of UTF-8 characters instead of single-by
+Line 745 
 of UTF-8 characters instead of single-by
  available only when PCRE is built to include UTF-8 support. If not, the use
  of this option provokes an error. Details of how this option changes the
  behaviour of PCRE are given in the
- .\" HTML <a href="pcre.html#utf8support">
+ .\" HREF
+ \fBpcreunicode\fP
+ .\"
+ page.
+ .sp
+   PCRE_NO_UTF8_CHECK
+ .sp
+ When PCRE_UTF8 is set, the validity of the pattern as a UTF-8 string is
+ automatically checked. There is a discussion about the
+ .\" HTML <a href="pcre.html#utf8strings">
  .\" </a>
- section on UTF-8 support
+ validity of UTF-8 strings
  .\"
  in the main
  .\" HREF
  \fBpcre\fP
  .\"
- page.
+ page. If an invalid UTF-8 sequence of bytes is found, \fBpcre_compile()\fP
+ returns an error. If you already know that your pattern is valid, and you want
+ to skip this check for performance reasons, you can set the PCRE_NO_UTF8_CHECK
+ option. When it is set, the effect of passing an invalid UTF-8 string as a
+ pattern is undefined. It may cause your program to crash. Note that this option
+ can also be passed to \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP, to suppress
+ the UTF-8 validity checking of subject strings.
+ .
+ .
+ .SH "COMPILATION ERROR CODES"
+ .rs
  .sp
-   PCRE_NO_UTF8_CHECK
+ The following table lists the error codes than may be returned by
+ \fBpcre_compile2()\fP, along with the error messages that may be returned by
+ both compiling functions. As PCRE has developed, some error codes have fallen
+ out of use. To avoid confusion, they have not been re-used.
+ .sp
+ no error
+ \e at end of pattern
+ \ec at end of pattern
+ unrecognized character follows \e
+ numbers out of order in {} quantifier
+ number too big in {} quantifier
+ missing terminating ] for character class
+ invalid escape sequence in character class
+ range out of order in character class
+ nothing to repeat
+ [this code is not in use]
+ internal error: unexpected repeat
+ unrecognized character after (? or (?-
+ POSIX named classes are supported only within a class
+ missing )
+ reference to non-existent subpattern
+ erroffset passed as NULL
+ unknown option bit(s) set
+ missing ) after comment
+ [this code is not in use]
+ regular expression is too large
+ failed to get memory
+ unmatched parentheses
+ internal error: code overflow
+ unrecognized character after (?<
+ lookbehind assertion is not fixed length
+ malformed number or name after (?(
+ conditional group contains more than two branches
+ assertion expected after (?(
+ (?R or (?[+-]digits must be followed by )
+ unknown POSIX class name
+ POSIX collating elements are not supported
+ this version of PCRE is not compiled with PCRE_UTF8 support
+ [this code is not in use]
+ character value in \ex{...} sequence is too large
+ invalid condition (?(0)
+ \eC not allowed in lookbehind assertion
+ PCRE does not support \eL, \el, \eN{name}, \eU, or \eu
+ number after (?C is > 255
+ closing ) for (?C expected
+ recursive call could loop indefinitely
+ unrecognized character after (?P
+ syntax error in subpattern name (missing terminator)
+ two named subpatterns have the same name
+ invalid UTF-8 string
+ support for \eP, \ep, and \eX has not been compiled
+ malformed \eP or \ep sequence
+ unknown property name after \eP or \ep
+ subpattern name is too long (maximum 32 characters)
+ too many named subpatterns (maximum 10000)
+ [this code is not in use]
+ octal value is greater than \e377 (not in UTF-8 mode)
+ internal error: overran compiling workspace
+ internal error: previously-checked referenced subpattern
+         not found
+ DEFINE group contains more than one branch
+ repeating a DEFINE group is not allowed
+ inconsistent NEWLINE options
+ \eg is not followed by a braced, angle-bracketed, or quoted
+         name/number or by a plain number
+ a numbered reference must not be zero
+ an argument is not allowed for (*ACCEPT), (*FAIL), or (*COMMIT)
+ (*VERB) not recognized
+ number is too big
+ subpattern name expected
+ digit expected after (?+
+ ] is an invalid data character in JavaScript compatibility mode
+ different names for subpatterns of the same number are
+         not allowed
+ (*MARK) must have an argument
+ this version of PCRE is not compiled with PCRE_UCP support
+ \ec must be followed by an ASCII character
+ \ek is not followed by a braced, angle-bracketed, or quoted name
  .sp
- When PCRE_UTF8 is set, the validity of the pattern as a UTF-8 string is
+ The numbers 32 and 10000 in errors 48 and 49 are defaults; different values may
- automatically checked. If an invalid UTF-8 sequence of bytes is found,
+ be used if the limits were changed when PCRE was built.
- \fBpcre_compile()\fP returns an error. If you already know that your pattern is
- valid, and you want to skip this check for performance reasons, you can set the
- PCRE_NO_UTF8_CHECK option. When it is set, the effect of passing an invalid
- UTF-8 string as a pattern is undefined. It may cause your program to crash.
- Note that this option can also be passed to \fBpcre_exec()\fP, to suppress the
- UTF-8 validity checking of subject strings.
  .
  .
+ .\" HTML <a name="studyingapattern"></a>
  .SH "STUDYING A PATTERN"
  .rs
  .sp
- .B pcre_extra *pcre_study(const pcre *\fIcode\fP, int \fIoptions\fP,
+ .B pcre_extra *pcre_study(const pcre *\fIcode\fP, int \fIoptions\fP
  .ti +5n
  .B const char **\fIerrptr\fP);
  .PP
-Line 483 
 help speed up matching, \fBpcre_study()\
+Line 874 
 help speed up matching, \fBpcre_study()\
  results of the study.
  .P
  The returned value from \fBpcre_study()\fP can be passed directly to
- \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
- 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
- described
+ passed; these are described
  .\" HTML <a href="#extradata">
  .\" </a>
  below
  .\"
  in the section on matching a pattern.
  .P
- If studying the pattern does not produce any additional information,
+ If studying the pattern does not produce any useful information,
  \fBpcre_study()\fP returns NULL. In that circumstance, if the calling program
- wants to pass any of the other fields to \fBpcre_exec()\fP, it must set up its
+ wants to pass any of the other fields to \fBpcre_exec()\fP or
- own \fBpcre_extra\fP block.
+ \fBpcre_dfa_exec()\fP, it must set up its own \fBpcre_extra\fP block.
  .P
- The second argument of \fBpcre_study()\fP contains option bits. At present, no
+ The second argument of \fBpcre_study()\fP contains option bits. There is only
- options are defined, and this argument should always be zero.
+ one option: PCRE_STUDY_JIT_COMPILE. If this is set, and the just-in-time
+ compiler is available, the pattern is further compiled into machine code that
+ executes much faster than the \fBpcre_exec()\fP matching function. If
+ the just-in-time compiler is not available, this option is ignored. All other
+ bits in the \fIoptions\fP argument must be zero.
+ .P
+ JIT compilation is a heavyweight optimization. It can take some time for
+ patterns to be analyzed, and for one-off matches and simple patterns the
+ benefit of faster execution might be offset by a much slower study time.
+ Not all patterns can be optimized by the JIT compiler. For those that cannot be
+ handled, matching automatically falls back to the \fBpcre_exec()\fP
+ interpreter. For more details, see the
+ .\" HREF
+ \fBpcrejit\fP
+ .\"
+ documentation.
  .P
  The third argument for \fBpcre_study()\fP is a pointer for an error message. If
  studying succeeds (even if no data is returned), the variable it points to is
- set to NULL. Otherwise it points to a textual error message. You should
+ set to NULL. Otherwise it is set to point to a textual error message. This is a
- therefore test the error pointer for NULL after calling \fBpcre_study()\fP, to
+ static string that is part of the library. You must not try to free it. You
- be sure that it has run successfully.
+ should test the error pointer for NULL after calling \fBpcre_study()\fP, to be
+ sure that it has run successfully.
+ .P
+ When you are finished with a pattern, you can free the memory used for the
+ study data by calling \fBpcre_free_study()\fP. This function was added to the
+ API for release 8.20. For earlier versions, the memory could be freed with
+ \fBpcre_free()\fP, just like the pattern itself. This will still work in cases
+ where PCRE_STUDY_JIT_COMPILE is not used, but it is advisable to change to the
+ new function when convenient.
  .P
- This is a typical call to \fBpcre_study\fP():
+ This is a typical way in which \fBpcre_study\fP() is used (except that in a
+ real application there should be tests for errors):
  .sp
-   pcre_extra *pe;
+   int rc;
-   pe = pcre_study(
+   pcre *re;
+   pcre_extra *sd;
+   re = pcre_compile("pattern", 0, &error, &erroroffset, NULL);
+   sd = pcre_study(
      re,             /* result of pcre_compile() */
-,              /* no options exist */
+,              /* no options */
      &error);        /* set to NULL or points to a message */
- .sp
+   rc = pcre_exec(   /* see below for details of pcre_exec() options */
- At present, studying a pattern is useful only for non-anchored patterns that do
+     re, sd, "subject", 7, 0, 0, ovector, 30);
- not have a single fixed starting character. A bitmap of possible starting
+   ...
- bytes is created.
+   pcre_free_study(sd);
+   pcre_free(re);
+ .sp
+ Studying a pattern does two things: first, a lower bound for the length of
+ subject string that is needed to match the pattern is computed. This does not
+ mean that there are any strings of that length that match, but it does
+ guarantee that no shorter strings match. The value is used by
+ \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP to avoid wasting time by trying to
+ match strings that are shorter than the lower bound. You can find out the value
+ in a calling program via the \fBpcre_fullinfo()\fP function.
+ .P
+ Studying a pattern is also useful for non-anchored patterns that do not have a
+ single fixed starting character. A bitmap of possible starting bytes is
+ created. This speeds up finding a position in the subject at which to start
+ matching.
+ .P
+ These two optimizations apply to both \fBpcre_exec()\fP and
+ \fBpcre_dfa_exec()\fP. However, they are not used by \fBpcre_exec()\fP if
+ \fBpcre_study()\fP is called with the PCRE_STUDY_JIT_COMPILE option, and
+ just-in-time compiling is successful. The optimizations can be disabled by
+ setting the PCRE_NO_START_OPTIMIZE option when calling \fBpcre_exec()\fP or
+ \fBpcre_dfa_exec()\fP. You might want to do this if your pattern contains
+ callouts or (*MARK) (which cannot be handled by the JIT compiler), and you want
+ to make use of these facilities in cases where matching fails. See the
+ discussion of PCRE_NO_START_OPTIMIZE
+ .\" HTML <a href="#execoptions">
+ .\" </a>
+ below.
+ .\"
  .
  .
  .\" HTML <a name="localesupport"></a>
-Line 525 
 bytes is created.
+Line 971 
 bytes is created.
  .sp
  PCRE handles caseless matching, and determines whether characters are letters,
  digits, or whatever, by reference to a set of tables, indexed by character
- value. (When running in UTF-8 mode, this applies only to characters with codes
+ value. When running in UTF-8 mode, this applies only to characters with codes
- less than 128. Higher-valued codes never match escapes such as \ew or \ed, but
+ less than 128. By default, higher-valued codes never match escapes such as \ew
- can be tested with \ep if PCRE is built with Unicode character property
+ or \ed, but they can be tested with \ep if PCRE is built with Unicode character
- support.)
+ property support. Alternatively, the PCRE_UCP option can be set at compile
- .P
+ time; this causes \ew and friends to use Unicode property support instead of
- An internal set of tables is created in the default C locale when PCRE is
+ built-in tables. The use of locales with Unicode is discouraged. If you are
- built. This is used when the final argument of \fBpcre_compile()\fP is NULL,
+ handling characters with codes greater than 128, you should either use UTF-8
- and is sufficient for many applications. An alternative set of tables can,
+ and Unicode, or use locales, but not try to mix the two.
- however, be supplied. These may be created in a different locale from the
+ .P
- default. As more and more applications change to using Unicode, the need for
+ PCRE contains an internal set of tables that are used when the final argument
- this locale support is expected to die away.
+ of \fBpcre_compile()\fP is NULL. These are sufficient for many applications.
+ Normally, the internal tables recognize only ASCII characters. However, when
+ PCRE is built, it is possible to cause the internal tables to be rebuilt in the
+ default "C" locale of the local system, which may cause them to be different.
+ .P
+ The internal tables can always be overridden by tables supplied by the
+ application that calls PCRE. These may be created in a different locale from
+ the default. As more and more applications change to using Unicode, the need
+ for this locale support is expected to die away.
  .P
  External tables are built by calling the \fBpcre_maketables()\fP function,
  which has no arguments, in the relevant locale. The result can then be passed
-Line 548 
 the following code could be used:
+Line 1002 
 the following code could be used:
    tables = pcre_maketables();
    re = pcre_compile(..., tables);
  .sp
+ The locale name "fr_FR" is used on Linux and other Unix-like systems; if you
+ are using Windows, the name for the French locale is "french".
+ .P
  When \fBpcre_maketables()\fP runs, the tables are built in memory that is
  obtained via \fBpcre_malloc\fP. It is the caller's responsibility to ensure
  that the memory containing the tables remains available for as long as it is
-Line 566 
 one in which it was compiled. Passing ta
+Line 1023 
 one in which it was compiled. Passing ta
  below in the section on matching a pattern.
  .
  .
+ .\" HTML <a name="infoaboutpattern"></a>
  .SH "INFORMATION ABOUT A PATTERN"
  .rs
  .sp
-Line 594 
 check against passing an arbitrary memor
+Line 1052 
 check against passing an arbitrary memor
  \fBpcre_fullinfo()\fP, to obtain the length of the compiled pattern:
  .sp
    int rc;
-   unsigned long int length;
+   size_t length;
    rc = pcre_fullinfo(
      re,               /* result of pcre_compile() */
-     pe,               /* result of pcre_study(), or NULL */
+     sd,               /* result of pcre_study(), or NULL */
      PCRE_INFO_SIZE,   /* what is required */
      &length);         /* where to put the data */
  .sp
-Line 615 
 no back references.
+Line 1073 
 no back references.
  Return the number of capturing subpatterns in the pattern. The fourth argument
  should point to an \fBint\fP variable.
  .sp
-   PCRE_INFO_DEFAULTTABLES
+   PCRE_INFO_DEFAULT_TABLES
  .sp
  Return a pointer to the internal default character tables within PCRE. The
  fourth argument should point to an \fBunsigned char *\fP variable. This
-Line 626 
 a NULL table pointer.
+Line 1084 
 a NULL table pointer.
    PCRE_INFO_FIRSTBYTE
  .sp
  Return information about the first byte of any matched string, for a
- non-anchored pattern. (This option used to be called PCRE_INFO_FIRSTCHAR; the
+ non-anchored pattern. The fourth argument should point to an \fBint\fP
- old name is still recognized for backwards compatibility.)
+ variable. (This option used to be called PCRE_INFO_FIRSTCHAR; the old name is
+ still recognized for backwards compatibility.)
  .P
  If there is a fixed first byte, for example, from a pattern such as
- (cat|cow|coyote), it is returned in the integer pointed to by \fIwhere\fP.
+ (cat|cow|coyote), its value is returned. Otherwise, if either
- Otherwise, if either
  .sp
  (a) the pattern was compiled with the PCRE_MULTILINE option, and every branch
  starts with "^", or
-Line 650 
 table indicating a fixed set of bytes fo
+Line 1108 
 table indicating a fixed set of bytes fo
  string, a pointer to the table is returned. Otherwise NULL is returned. The
  fourth argument should point to an \fBunsigned char *\fP variable.
  .sp
+   PCRE_INFO_HASCRORLF
+ .sp
+ Return 1 if the pattern contains any explicit matches for CR or LF characters,
+ otherwise 0. The fourth argument should point to an \fBint\fP variable. An
+ explicit match is either a literal CR or LF character, or \er or \en.
+ .sp
+   PCRE_INFO_JCHANGED
+ .sp
+ Return 1 if the (?J) or (?-J) option setting is used in the pattern, otherwise
+. The fourth argument should point to an \fBint\fP variable. (?J) and
+ (?-J) set and unset the local PCRE_DUPNAMES option, respectively.
+ .sp
+   PCRE_INFO_JIT
+ .sp
+ Return 1 if the pattern was studied with the PCRE_STUDY_JIT_COMPILE option, and
+ just-in-time compiling was successful. The fourth argument should point to an
+ \fBint\fP variable. A return value of 0 means that JIT support is not available
+ in this version of PCRE, or that the pattern was not studied with the
+ PCRE_STUDY_JIT_COMPILE option, or that the JIT compiler could not handle this
+ particular pattern. See the
+ .\" HREF
+ \fBpcrejit\fP
+ .\"
+ documentation for details of what can and cannot be handled.
+ .sp
    PCRE_INFO_LASTLITERAL
  .sp
  Return the value of the rightmost literal byte that must exist in any matched
-Line 660 
 follows something of variable length. Fo
+Line 1143 
 follows something of variable length. Fo
  /^a\ed+z\ed+/ the returned value is "z", but for /^a\edz\ed/ the returned value
  is -1.
  .sp
+   PCRE_INFO_MINLENGTH
+ .sp
+ If the pattern was studied and a minimum length for matching subject strings
+ was computed, its value is returned. Otherwise the returned value is -1. The
+ value is a number of characters, not bytes (this may be relevant in UTF-8
+ mode). The fourth argument should point to an \fBint\fP variable. A
+ non-negative value is a lower bound to the length of any matching string. There
+ may not be any strings of that length that do actually match, but every string
+ that does match is at least that long.
+ .sp
    PCRE_INFO_NAMECOUNT
    PCRE_INFO_NAMEENTRYSIZE
    PCRE_INFO_NAMETABLE
  .sp
  PCRE supports the use of named as well as numbered capturing parentheses. The
  names are just an additional way of identifying the parentheses, which still
- acquire numbers. A convenience function called \fBpcre_get_named_substring()\fP
+ acquire numbers. Several convenience functions such as
- is provided for extracting an individual captured substring by name. It is also
+ \fBpcre_get_named_substring()\fP are provided for extracting captured
- possible to extract the data directly, by first converting the name to a number
+ substrings by name. It is also possible to extract the data directly, by first
- in order to access the correct pointers in the output vector (described with
+ converting the name to a number in order to access the correct pointers in the
- \fBpcre_exec()\fP below). To do the conversion, you need to use the
+ output vector (described with \fBpcre_exec()\fP below). To do the conversion,
- name-to-number map, which is described by these three values.
+ you need to use the name-to-number map, which is described by these three
+ values.
  .P
  The map consists of a number of fixed-size entries. PCRE_INFO_NAMECOUNT gives
  the number of entries, and PCRE_INFO_NAMEENTRYSIZE gives the size of each
-Line 679 
 entry; both of these return an \fBint\fP
+Line 1173 
 entry; both of these return an \fBint\fP
  length of the longest name. PCRE_INFO_NAMETABLE returns a pointer to the first
  entry of the table (a pointer to \fBchar\fP). The first two bytes of each entry
  are the number of the capturing parenthesis, most significant byte first. The
- rest of the entry is the corresponding name, zero terminated. The names are in
+ rest of the entry is the corresponding name, zero terminated.
- alphabetical order. For example, consider the following pattern (assume
+ .P
- PCRE_EXTENDED is set, so white space - including newlines - is ignored):
+ The names are in alphabetical order. Duplicate names may appear if (?| is used
+ to create multiple groups with the same number, as described in the
+ .\" HTML <a href="pcrepattern.html#dupsubpatternnumber">
+ .\" </a>
+ section on duplicate subpattern numbers
+ .\"
+ in the
+ .\" HREF
+ \fBpcrepattern\fP
+ .\"
+ page. Duplicate names for subpatterns with different numbers are permitted only
+ if PCRE_DUPNAMES is set. In all cases of duplicate names, they appear in the
+ table in the order in which they were found in the pattern. In the absence of
+ (?| this is the order of increasing number; when (?| is used this is not
+ necessarily the case because later subpatterns may have lower numbers.
+ .P
+ As a simple example of the name/number table, consider the following pattern
+ (assume PCRE_EXTENDED is set, so white space - including newlines - is
+ ignored):
  .sp
  .\" JOIN
-   (?P<date> (?P<year>(\ed\ed)?\ed\ed) -
+   (?<date> (?<year>(\ed\ed)?\ed\ed) -
-   (?P<month>\ed\ed) - (?P<day>\ed\ed) )
+   (?<month>\ed\ed) - (?<day>\ed\ed) )
  .sp
  There are four named subpatterns, so the table has four entries, and each entry
  in the table is eight bytes long. The table is as follows, with non-printing
-Line 697 
 bytes shows in hexadecimal, and undefine
+Line 1209 
 bytes shows in hexadecimal, and undefine
 02 y  e  a  r  00 ??
  .sp
  When writing code to extract data from named subpatterns using the
- name-to-number map, remember that the length of each entry is likely to be
+ name-to-number map, remember that the length of the entries is likely to be
  different for each compiled pattern.
  .sp
+   PCRE_INFO_OKPARTIAL
+ .sp
+ Return 1 if the pattern can be used for partial matching with
+ \fBpcre_exec()\fP, otherwise 0. The fourth argument should point to an
+ \fBint\fP variable. From release 8.00, this always returns 1, because the
+ restrictions that previously applied to partial matching have been lifted. The
+ .\" HREF
+ \fBpcrepartial\fP
+ .\"
+ documentation gives details of partial matching.
+ .sp
    PCRE_INFO_OPTIONS
  .sp
  Return a copy of the options with which the pattern was compiled. The fourth
  argument should point to an \fBunsigned long int\fP variable. These option bits
  are those specified in the call to \fBpcre_compile()\fP, modified by any
- top-level option settings within the pattern itself.
+ top-level option settings at the start of the pattern itself. In other words,
+ they are the options that will be in force when matching starts. For example,
+ if the pattern /(?im)abc(?-i)d/ is compiled with the PCRE_EXTENDED option, the
+ result is PCRE_CASELESS, PCRE_MULTILINE, and PCRE_EXTENDED.
  .P
  A pattern is automatically anchored by PCRE if all of its top-level
  alternatives begin with one of the following:
-Line 729 
 variable.
+Line 1255 
 variable.
  .sp
    PCRE_INFO_STUDYSIZE
  .sp
- Return the size of the data block pointed to by the \fIstudy_data\fP field in
+ Return the size of the data block pointed to by the \fIstudy_data\fP field in a
- a \fBpcre_extra\fP block. That is, it is the value that was passed to
+ \fBpcre_extra\fP block. If \fBpcre_extra\fP is NULL, or there is no study data,
- \fBpcre_malloc()\fP when PCRE was getting memory into which to place the data
+ zero is returned. The fourth argument should point to a \fBsize_t\fP variable.
- created by \fBpcre_study()\fP. The fourth argument should point to a
+ The \fIstudy_data\fP field is set by \fBpcre_study()\fP to record information
- \fBsize_t\fP variable.
+ that will speed up matching (see the section entitled
+ .\" HTML <a href="#studyingapattern">
+ .\" </a>
+ "Studying a pattern"
+ .\"
+ above). The format of the \fIstudy_data\fP block is private, but its length
+ is made available via this option so that it can be saved and restored (see the
+ .\" HREF
+ \fBpcreprecompile\fP
+ .\"
+ documentation for details).
  .
  .
  .SH "OBSOLETE INFO FUNCTION"
-Line 760 
 it is used to pass back information abou
+Line 1296 
 it is used to pass back information abou
  string (see PCRE_INFO_FIRSTBYTE above).
  .
  .
- .SH "MATCHING A PATTERN"
+ .SH "REFERENCE COUNTS"
+ .rs
+ .sp
+ .B int pcre_refcount(pcre *\fIcode\fP, int \fIadjust\fP);
+ .PP
+ The \fBpcre_refcount()\fP function is used to maintain a reference count in the
+ data block that contains a compiled pattern. It is provided for the benefit of
+ applications that operate in an object-oriented manner, where different parts
+ of the application may be using the same compiled pattern, but you want to free
+ the block when they are all done.
+ .P
+ When a pattern is compiled, the reference count field is initialized to zero.
+ It is changed only by calling this function, whose action is to add the
+ \fIadjust\fP value (which may be positive or negative) to it. The yield of the
+ function is the new value. However, the value of the count is constrained to
+ lie between 0 and 65535, inclusive. If the new value is outside these limits,
+ it is forced to the appropriate limit value.
+ .P
+ Except when it is zero, the reference count is not correctly preserved if a
+ pattern is compiled on one host and then transferred to a host whose byte-order
+ is different. (This seems a highly unlikely scenario.)
+ .
+ .
+ .SH "MATCHING A PATTERN: THE TRADITIONAL FUNCTION"
  .rs
  .sp
  .B int pcre_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
-Line 771 
 string (see PCRE_INFO_FIRSTBYTE above).
+Line 1330 
 string (see PCRE_INFO_FIRSTBYTE above).
  .P
  The function \fBpcre_exec()\fP is called to match a subject string against a
  compiled pattern, which is passed in the \fIcode\fP argument. If the
- 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
- \fIextra\fP argument.
+ \fIextra\fP argument. This function is the main matching facility of the
+ library, and it operates in a Perl-like manner. For specialist use there is
+ also an alternative matching function, which is described
+ .\" HTML <a href="#dfamatch">
+ .\" </a>
+ below
+ .\"
+ in the section about the \fBpcre_dfa_exec()\fP function.
  .P
  In most applications, the pattern will have been compiled (and optionally
  studied) in the same process that calls \fBpcre_exec()\fP. However, it is
-Line 796 
 Here is an example of a simple call to \
+Line 1362 
 Here is an example of a simple call to \
 ,              /* start at offset 0 in the subject */
 ,              /* default options */
      ovector,        /* vector of integers for substring information */
-);            /* number of elements in the vector (NOT size in bytes) */
+);            /* number of elements (NOT size in bytes) */
+ .
  .
  .\" HTML <a name="extradata"></a>
  .SS "Extra data for \fBpcre_exec()\fR"
-Line 805 
 Here is an example of a simple call to \
+Line 1372 
 Here is an example of a simple call to \
  If the \fIextra\fP argument is not NULL, it must point to a \fBpcre_extra\fP
  data block. The \fBpcre_study()\fP function returns such a block (when it
  doesn't return NULL), but you can also create one for yourself, and pass
- additional information in it. The fields in a \fBpcre_extra\fP block are as
+ additional information in it. The \fBpcre_extra\fP block contains the following
- follows:
+ fields (not necessarily in this order):
  .sp
    unsigned long int \fIflags\fP;
    void *\fIstudy_data\fP;
+   void *\fIexecutable_jit\fP;
    unsigned long int \fImatch_limit\fP;
+   unsigned long int \fImatch_limit_recursion\fP;
    void *\fIcallout_data\fP;
    const unsigned char *\fItables\fP;
+   unsigned char **\fImark\fP;
  .sp
  The \fIflags\fP field is a bitmap that specifies which of the other fields
  are set. The flag bits are:
  .sp
    PCRE_EXTRA_STUDY_DATA
+   PCRE_EXTRA_EXECUTABLE_JIT
    PCRE_EXTRA_MATCH_LIMIT
+   PCRE_EXTRA_MATCH_LIMIT_RECURSION
    PCRE_EXTRA_CALLOUT_DATA
    PCRE_EXTRA_TABLES
+   PCRE_EXTRA_MARK
  .sp
- 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
- \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
- 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
- the block by setting the other fields and their corresponding flag bits.
+ should not set these yourself, but you may add to the block by setting the
+ other fields and their corresponding flag bits.
  .P
  The \fImatch_limit\fP field provides a means of preventing PCRE from using up a
  vast amount of resources when running patterns that are not going to match,
  but which have a very large number of possibilities in their search trees. The
- classic example is the use of nested unlimited repeats.
+ classic example is a pattern that uses nested unlimited repeats.
  .P
- Internally, PCRE uses a function called \fBmatch()\fP which it calls repeatedly
+ Internally, \fBpcre_exec()\fP uses a function called \fBmatch()\fP, which it
- (sometimes recursively). The limit is imposed on the number of times this
+ calls repeatedly (sometimes recursively). The limit set by \fImatch_limit\fP is
- function is called during a match, which has the effect of limiting the amount
+ imposed on the number of times this function is called during a match, which
- of recursion and backtracking that can take place. For patterns that are not
+ has the effect of limiting the amount of backtracking that can take place. For
- anchored, the count starts from zero for each position in the subject string.
+ patterns that are not anchored, the count restarts from zero for each position
+ in the subject string.
+ .P
+ When \fBpcre_exec()\fP is called with a pattern that was successfully studied
+ with the PCRE_STUDY_JIT_COMPILE option, the way that the matching is executed
+ is entirely different. However, there is still the possibility of runaway
+ matching that goes on for a very long time, and so the \fImatch_limit\fP value
+ is also used in this case (but in a different way) to limit how long the
+ matching can continue.
  .P
- The default limit for the library can be set when PCRE is built; the default
+ The default value for the limit can be set when PCRE is built; the default
  default is 10 million, which handles all but the most extreme cases. You can
- reduce the default by suppling \fBpcre_exec()\fP with a \fBpcre_extra\fP block
+ override the default by suppling \fBpcre_exec()\fP with a \fBpcre_extra\fP
- in which \fImatch_limit\fP is set to a smaller value, and
+ block in which \fImatch_limit\fP is set, and PCRE_EXTRA_MATCH_LIMIT is set in
- PCRE_EXTRA_MATCH_LIMIT is set in the \fIflags\fP field. If the limit is
+ the \fIflags\fP field. If the limit is exceeded, \fBpcre_exec()\fP returns
- exceeded, \fBpcre_exec()\fP returns PCRE_ERROR_MATCHLIMIT.
+ PCRE_ERROR_MATCHLIMIT.
+ .P
+ The \fImatch_limit_recursion\fP field is similar to \fImatch_limit\fP, but
+ instead of limiting the total number of times that \fBmatch()\fP is called, it
+ limits the depth of recursion. The recursion depth is a smaller number than the
+ total number of calls, because not all calls to \fBmatch()\fP are recursive.
+ This limit is of use only if it is set smaller than \fImatch_limit\fP.
+ .P
+ Limiting the recursion depth limits the amount of machine stack that can be
+ used, or, when PCRE has been compiled to use memory on the heap instead of the
+ stack, the amount of heap memory that can be used. This limit is not relevant,
+ and is ignored, if the pattern was successfully studied with
+ PCRE_STUDY_JIT_COMPILE.
+ .P
+ The default value for \fImatch_limit_recursion\fP can be set when PCRE is
+ built; the default default is the same value as the default for
+ \fImatch_limit\fP. You can override the default by suppling \fBpcre_exec()\fP
+ with a \fBpcre_extra\fP block in which \fImatch_limit_recursion\fP is set, and
+ PCRE_EXTRA_MATCH_LIMIT_RECURSION is set in the \fIflags\fP field. If the limit
+ is exceeded, \fBpcre_exec()\fP returns PCRE_ERROR_RECURSIONLIMIT.
  .P
- The \fIpcre_callout\fP field is used in conjunction with the "callout" feature,
+ The \fIcallout_data\fP field is used in conjunction with the "callout" feature,
- which is described in the
+ and is described in the
  .\" HREF
  \fBpcrecallout\fP
  .\"
-Line 865 
 called. See the
+Line 1466 
 called. See the
  \fBpcreprecompile\fP
  .\"
  documentation for a discussion of saving compiled patterns for later use.
+ .P
+ If PCRE_EXTRA_MARK is set in the \fIflags\fP field, the \fImark\fP field must
+ be set to point to a \fBchar *\fP variable. If the pattern contains any
+ backtracking control verbs such as (*MARK:NAME), and the execution ends up with
+ a name to pass back, a pointer to the name string (zero terminated) is placed
+ in the variable pointed to by the \fImark\fP field. The names are within the
+ compiled pattern; if you wish to retain such a name you must copy it before
+ freeing the memory of a compiled pattern. If there is no name to pass back, the
+ variable pointed to by the \fImark\fP field set to NULL. For details of the
+ backtracking control verbs, see the section entitled
+ .\" HTML <a href="pcrepattern#backtrackcontrol">
+ .\" </a>
+ "Backtracking control"
+ .\"
+ in the
+ .\" HREF
+ \fBpcrepattern\fP
+ .\"
+ documentation.
+ .
  .
+ .\" HTML <a name="execoptions"></a>
  .SS "Option bits for \fBpcre_exec()\fP"
  .rs
  .sp
  The unused bits of the \fIoptions\fP argument for \fBpcre_exec()\fP must be
- zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NOTBOL,
+ zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_\fIxxx\fP,
- PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NO_UTF8_CHECK and PCRE_PARTIAL.
+ PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NOTEMPTY_ATSTART,
+ PCRE_NO_START_OPTIMIZE, PCRE_NO_UTF8_CHECK, PCRE_PARTIAL_SOFT, and
+ PCRE_PARTIAL_HARD.
+ .P
+ If the pattern was successfully studied with the PCRE_STUDY_JIT_COMPILE option,
+ the only supported options for JIT execution are PCRE_NO_UTF8_CHECK,
+ PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, and PCRE_NOTEMPTY_ATSTART. Note in
+ particular that partial matching is not supported. If an unsupported option is
+ used, JIT execution is disabled and the normal interpretive code in
+ \fBpcre_exec()\fP is run.
  .sp
    PCRE_ANCHORED
  .sp
-Line 880 
 matching position. If a pattern was comp
+Line 1511 
 matching position. If a pattern was comp
  to be anchored by virtue of its contents, it cannot be made unachored at
  matching time.
  .sp
+   PCRE_BSR_ANYCRLF
+   PCRE_BSR_UNICODE
+ .sp
+ These options (which are mutually exclusive) control what the \eR escape
+ sequence matches. The choice is either to match only CR, LF, or CRLF, or to
+ match any Unicode newline sequence. These options override the choice that was
+ made or defaulted when the pattern was compiled.
+ .sp
+   PCRE_NEWLINE_CR
+   PCRE_NEWLINE_LF
+   PCRE_NEWLINE_CRLF
+   PCRE_NEWLINE_ANYCRLF
+   PCRE_NEWLINE_ANY
+ .sp
+ These options override the newline definition that was chosen or defaulted when
+ the pattern was compiled. For details, see the description of
+ \fBpcre_compile()\fP above. During matching, the newline choice affects the
+ behaviour of the dot, circumflex, and dollar metacharacters. It may also alter
+ the way the match position is advanced after a match failure for an unanchored
+ pattern.
+ .P
+ When PCRE_NEWLINE_CRLF, PCRE_NEWLINE_ANYCRLF, or PCRE_NEWLINE_ANY is set, and a
+ match attempt for an unanchored pattern fails when the current position is at a
+ CRLF sequence, and the pattern contains no explicit matches for CR or LF
+ characters, the match position is advanced by two characters instead of one, in
+ other words, to after the CRLF.
+ .P
+ The above rule is a compromise that makes the most common cases work as
+ expected. For example, if the pattern is .+A (and the PCRE_DOTALL option is not
+ set), it does not match the string "\er\enA" because, after failing at the
+ start, it skips both the CR and the LF before retrying. However, the pattern
+ [\er\en]A does match that string, because it contains an explicit CR or LF
+ reference, and so advances only by one character after the first failure.
+ .P
+ An explicit match for CR of LF is either a literal appearance of one of those
+ characters, or one of the \er or \en escape sequences. Implicit matches such as
+ [^X] do not count, nor does \es (which includes CR and LF in the characters
+ that it matches).
+ .P
+ Notwithstanding the above, anomalous effects may still occur when CRLF is a
+ valid newline sequence and explicit \er or \en escapes appear in the pattern.
+ .sp
    PCRE_NOTBOL
  .sp
  This option specifies that first character of the subject string is not the
-Line 904 
 match the empty string, the entire match
+Line 1577 
 match the empty string, the entire match
  .sp
    a?b?
  .sp
- is applied to a string not beginning with "a" or "b", it matches the empty
+ is applied to a string not beginning with "a" or "b", it matches an empty
  string at the start of the subject. With PCRE_NOTEMPTY set, this match is not
  valid, so PCRE searches further into the string for occurrences of "a" or "b".
+ .sp
+   PCRE_NOTEMPTY_ATSTART
+ .sp
+ This is like PCRE_NOTEMPTY, except that an empty string match that is not at
+ the start of the subject is permitted. If the pattern is anchored, such a match
+ can occur only if the pattern contains \eK.
+ .P
+ Perl has no direct equivalent of PCRE_NOTEMPTY or PCRE_NOTEMPTY_ATSTART, but it
+ does make a special case of a pattern match of the empty string within its
+ \fBsplit()\fP function, and when using the /g modifier. It is possible to
+ emulate Perl's behaviour after matching a null string by first trying the match
+ again at the same offset with PCRE_NOTEMPTY_ATSTART and PCRE_ANCHORED, and then
+ if that fails, by advancing the starting offset (see below) and trying an
+ ordinary match again. There is some code that demonstrates how to do this in
+ the
+ .\" HREF
+ \fBpcredemo\fP
+ .\"
+ sample program. In the most general case, you have to check to see if the
+ newline convention recognizes CRLF as a newline, and if so, and the current
+ character is CR followed by LF, advance the starting offset by two characters
+ instead of one.
+ .sp
+   PCRE_NO_START_OPTIMIZE
+ .sp
+ There are a number of optimizations that \fBpcre_exec()\fP uses at the start of
+ a match, in order to speed up the process. For example, if it is known that an
+ unanchored match must start with a specific character, it searches the subject
+ for that character, and fails immediately if it cannot find it, without
+ actually running the main matching function. This means that a special item
+ such as (*COMMIT) at the start of a pattern is not considered until after a
+ suitable starting point for the match has been found. When callouts or (*MARK)
+ items are in use, these "start-up" optimizations can cause them to be skipped
+ if the pattern is never actually used. The start-up optimizations are in effect
+ a pre-scan of the subject that takes place before the pattern is run.
+ .P
+ The PCRE_NO_START_OPTIMIZE option disables the start-up optimizations, possibly
+ causing performance to suffer, but ensuring that in cases where the result is
+ "no match", the callouts do occur, and that items such as (*COMMIT) and (*MARK)
+ are considered at every possible starting position in the subject string. If
+ PCRE_NO_START_OPTIMIZE is set at compile time, it cannot be unset at matching
+ time.
  .P
- Perl has no direct equivalent of PCRE_NOTEMPTY, but it does make a special case
+ Setting PCRE_NO_START_OPTIMIZE can change the outcome of a matching operation.
- of a pattern match of the empty string within its \fBsplit()\fP function, and
+ Consider the pattern
- when using the /g modifier. It is possible to emulate Perl's behaviour after
+ .sp
- matching a null string by first trying the match again at the same offset with
+   (*COMMIT)ABC
- PCRE_NOTEMPTY and PCRE_ANCHORED, and then if that fails by advancing the
+ .sp
- starting offset (see below) and trying an ordinary match again. There is some
+ When this is compiled, PCRE records the fact that a match must start with the
- code that demonstrates how to do this in the \fIpcredemo.c\fP sample program.
+ character "A". Suppose the subject string is "DEFABC". The start-up
+ optimization scans along the subject, finds "A" and runs the first match
+ attempt from there. The (*COMMIT) item means that the pattern must match the
+ current starting position, which in this case, it does. However, if the same
+ match is run with PCRE_NO_START_OPTIMIZE set, the initial scan along the
+ subject string does not happen. The first match attempt is run starting from
+ "D" and when this fails, (*COMMIT) prevents any further matches being tried, so
+ the overall result is "no match". If the pattern is studied, more start-up
+ optimizations may be used. For example, a minimum length for the subject may be
+ recorded. Consider the pattern
+ .sp
+   (*MARK:A)(X|Y)
+ .sp
+ The minimum length for a match is one character. If the subject is "ABC", there
+ will be attempts to match "ABC", "BC", "C", and then finally an empty string.
+ If the pattern is studied, the final attempt does not take place, because PCRE
+ knows that the subject is too short, and so the (*MARK) is never encountered.
+ In this case, studying the pattern does not affect the overall match result,
+ which is still "no match", but it does affect the auxiliary information that is
+ returned.
  .sp
    PCRE_NO_UTF8_CHECK
  .sp
  When PCRE_UTF8 is set at compile time, the validity of the subject as a UTF-8
  string is automatically checked when \fBpcre_exec()\fP is subsequently called.
  The value of \fIstartoffset\fP is also checked to ensure that it points to the
- start of a UTF-8 character. If an invalid UTF-8 sequence of bytes is found,
+ start of a UTF-8 character. There is a discussion about the validity of UTF-8
- \fBpcre_exec()\fP returns the error PCRE_ERROR_BADUTF8. If \fIstartoffset\fP
+ strings in the
- contains an invalid value, PCRE_ERROR_BADUTF8_OFFSET is returned.
+ .\" HTML <a href="pcre.html#utf8strings">
+ .\" </a>
+ section on UTF-8 support
+ .\"
+ in the main
+ .\" HREF
+ \fBpcre\fP
+ .\"
+ page. If an invalid UTF-8 sequence of bytes is found, \fBpcre_exec()\fP returns
+ the error PCRE_ERROR_BADUTF8 or, if PCRE_PARTIAL_HARD is set and the problem is
+ a truncated UTF-8 character at the end of the subject, PCRE_ERROR_SHORTUTF8. In
+ both cases, information about the precise nature of the error may also be
+ returned (see the descriptions of these errors in the section entitled \fIError
+ return values from\fP \fBpcre_exec()\fP
+ .\" HTML <a href="#errorlist">
+ .\" </a>
+ below).
+ .\"
+ If \fIstartoffset\fP contains a value that does not point to the start of a
+ UTF-8 character (or to the end of the subject), PCRE_ERROR_BADUTF8_OFFSET is
+ returned.
  .P
  If you already know that your subject is valid, and you want to skip these
  checks for performance reasons, you can set the PCRE_NO_UTF8_CHECK option when
  calling \fBpcre_exec()\fP. You might want to do this for the second and
  subsequent calls to \fBpcre_exec()\fP if you are making repeated calls to find
  all the matches in a single subject string. However, you should be sure that
- 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 UTF-8 character (or the
- PCRE_NO_UTF8_CHECK is set, the effect of passing an invalid UTF-8 string as a
+ end of the subject). When PCRE_NO_UTF8_CHECK is set, the effect of passing an
- subject, or a value of \fIstartoffset\fP that does not point to the start of a
+ invalid UTF-8 string as a subject or an invalid value of \fIstartoffset\fP is
- UTF-8 character, is undefined. Your program may crash.
+ undefined. Your program may crash.
  .sp
-   PCRE_PARTIAL
+   PCRE_PARTIAL_HARD
- .sp
+   PCRE_PARTIAL_SOFT
- This option turns on the partial matching feature. If the subject string fails
+ .sp
- to match the pattern, but at some point during the matching process the end of
+ These options turn on the partial matching feature. For backwards
- the subject was reached (that is, the subject partially matches the pattern and
+ compatibility, PCRE_PARTIAL is a synonym for PCRE_PARTIAL_SOFT. A partial match
- the failure to match occurred only because there were not enough subject
+ occurs if the end of the subject string is reached successfully, but there are
- characters), \fBpcre_exec()\fP returns PCRE_ERROR_PARTIAL instead of
+ not enough subject characters to complete the match. If this happens when
- PCRE_ERROR_NOMATCH. When PCRE_PARTIAL is used, there are restrictions on what
+ PCRE_PARTIAL_SOFT (but not PCRE_PARTIAL_HARD) is set, matching continues by
- may appear in the pattern. These are discussed in the
+ testing any remaining alternatives. Only if no complete match can be found is
+ PCRE_ERROR_PARTIAL returned instead of PCRE_ERROR_NOMATCH. In other words,
+ PCRE_PARTIAL_SOFT says that the caller is prepared to handle a partial match,
+ but only if no complete match can be found.
+ .P
+ If PCRE_PARTIAL_HARD is set, it overrides PCRE_PARTIAL_SOFT. In this case, if a
+ partial match is found, \fBpcre_exec()\fP immediately returns
+ PCRE_ERROR_PARTIAL, without considering any other alternatives. In other words,
+ when PCRE_PARTIAL_HARD is set, a partial match is considered to be more
+ important that an alternative complete match.
+ .P
+ In both cases, the portion of the string that was inspected when the partial
+ match was found is set as the first matching string. There is a more detailed
+ discussion of partial and multi-segment matching, with examples, in the
  .\" HREF
  \fBpcrepartial\fP
  .\"
  documentation.
  .
+ .
  .SS "The string to be matched by \fBpcre_exec()\fP"
  .rs
  .sp
  The subject string is passed to \fBpcre_exec()\fP as a pointer in
- \fIsubject\fP, a length in \fIlength\fP, and a starting byte offset in
+ \fIsubject\fP, a length (in bytes) in \fIlength\fP, and a starting byte offset
- \fIstartoffset\fP. In UTF-8 mode, the byte offset must point to the start of a
+ in \fIstartoffset\fP. If this is negative or greater than the length of the
- UTF-8 character. Unlike the pattern string, the subject may contain binary zero
+ subject, \fBpcre_exec()\fP returns PCRE_ERROR_BADOFFSET. When the starting
- bytes. When the starting offset is zero, the search for a match starts at the
+ offset is zero, the search for a match starts at the beginning of the subject,
- 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
+ point to the start of a UTF-8 character (or the end of the subject). Unlike the
+ pattern string, the subject may contain binary zero bytes.
  .P
  A non-zero starting offset is useful when searching for another match in the
  same subject by calling \fBpcre_exec()\fP again after a previous success.
-Line 977 
 start of the subject, which is deemed to
+Line 1747 
 start of the subject, which is deemed to
  set to 4, it finds the second occurrence of "iss" because it is able to look
  behind the starting point to discover that it is preceded by a letter.
  .P
+ Finding all the matches in a subject is tricky when the pattern can match an
+ empty string. It is possible to emulate Perl's /g behaviour by first trying the
+ match again at the same offset, with the PCRE_NOTEMPTY_ATSTART and
+ PCRE_ANCHORED options, and then if that fails, advancing the starting offset
+ and trying an ordinary match again. There is some code that demonstrates how to
+ do this in the
+ .\" HREF
+ \fBpcredemo\fP
+ .\"
+ sample program. In the most general case, you have to check to see if the
+ newline convention recognizes CRLF as a newline, and if so, and the current
+ character is CR followed by LF, advance the starting offset by two characters
+ instead of one.
+ .P
  If a non-zero starting offset is passed when the pattern is anchored, one
  attempt to match at the given offset is made. This can only succeed if the
  pattern does not require the match to be at the start of the subject.
  .
+ .
  .SS "How \fBpcre_exec()\fP returns captured substrings"
  .rs
  .sp
-Line 991 
 pattern. Following the usage in Jeffrey
+Line 1776 
 pattern. Following the usage in Jeffrey
  a fragment of a pattern that picks out a substring. PCRE supports several other
  kinds of parenthesized subpattern that do not cause substrings to be captured.
  .P
- Captured substrings are returned to the caller via a vector of integer offsets
+ Captured substrings are returned to the caller via a vector of integers whose
- whose address is passed in \fIovector\fP. The number of elements in the vector
+ address is passed in \fIovector\fP. The number of elements in the vector is
- is passed in \fIovecsize\fP, which must be a non-negative number. \fBNote\fP:
+ passed in \fIovecsize\fP, which must be a non-negative number. \fBNote\fP: this
- this argument is NOT the size of \fIovector\fP in bytes.
+ argument is NOT the size of \fIovector\fP in bytes.
  .P
  The first two-thirds of the vector is used to pass back captured substrings,
  each substring using a pair of integers. The remaining third of the vector is
  used as workspace by \fBpcre_exec()\fP while matching capturing subpatterns,
- and is not available for passing back information. The length passed in
+ and is not available for passing back information. The number passed in
  \fIovecsize\fP should always be a multiple of three. If it is not, it is
  rounded down.
  .P
  When a match is successful, information about captured substrings is returned
  in pairs of integers, starting at the beginning of \fIovector\fP, and
- continuing up to two-thirds of its length at the most. The first element of a
+ continuing up to two-thirds of its length at the most. The first element of
- pair is set to the offset of the first character in a substring, and the second
+ each pair is set to the byte offset of the first character in a substring, and
- is set to the offset of the first character after the end of a substring. The
+ the second is set to the byte offset of the first character after the end of a
- first pair, \fIovector[0]\fP and \fIovector[1]\fP, identify the portion of the
+ substring. \fBNote\fP: these values are always byte offsets, even in UTF-8
- subject string matched by the entire pattern. The next pair is used for the
+ mode. They are not character counts.
- first capturing subpattern, and so on. The value returned by \fBpcre_exec()\fP
+ .P
- is the number of pairs that have been set. If there are no capturing
+ The first pair of integers, \fIovector[0]\fP and \fIovector[1]\fP, identify the
- subpatterns, the return value from a successful match is 1, indicating that
+ portion of the subject string matched by the entire pattern. The next pair is
- just the first pair of offsets has been set.
+ used for the first capturing subpattern, and so on. The value returned by
- .P
+ \fBpcre_exec()\fP is one more than the highest numbered pair that has been set.
- Some convenience functions are provided for extracting the captured substrings
+ For example, if two substrings have been captured, the returned value is 3. If
- as separate strings. These are described in the following section.
+ there are no capturing subpatterns, the return value from a successful match is
- .P
+, indicating that just the first pair of offsets has been set.
- It is possible for an capturing subpattern number \fIn+1\fP to match some
- part of the subject when subpattern \fIn\fP has not been used at all. For
- example, if the string "abc" is matched against the pattern (a|(z))(bc)
- subpatterns 1 and 3 are matched, but 2 is not. When this happens, both offset
- values corresponding to the unused subpattern are set to -1.
  .P
  If a capturing subpattern is matched repeatedly, it is the last portion of the
  string that it matched that is returned.
  .P
  If the vector is too small to hold all the captured substring offsets, it is
  used as far as possible (up to two-thirds of its length), and the function
- returns a value of zero. In particular, if the substring offsets are not of
+ returns a value of zero. If neither the actual string matched not any captured
- interest, \fBpcre_exec()\fP may be called with \fIovector\fP passed as NULL and
+ substrings are of interest, \fBpcre_exec()\fP may be called with \fIovector\fP
- \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
- 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
- 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
- advisable to supply an \fIovector\fP.
+ is usually advisable to supply an \fIovector\fP of reasonable size.
+ .P
+ There are some cases where zero is returned (indicating vector overflow) when
+ in fact the vector is exactly the right size for the final match. For example,
+ consider the pattern
+ .sp
+   (a)(?:(b)c|bd)
+ .sp
+ If a vector of 6 elements (allowing for only 1 captured substring) is given
+ with subject string "abd", \fBpcre_exec()\fP will try to set the second
+ captured string, thereby recording a vector overflow, before failing to match
+ "c" and backing up to try the second alternative. The zero return, however,
+ does correctly indicate that the maximum number of slots (namely 2) have been
+ filled. In similar cases where there is temporary overflow, but the final
+ number of used slots is actually less than the maximum, a non-zero value is
+ returned.
  .P
- Note that \fBpcre_info()\fP can be used to find out how many capturing
+ The \fBpcre_fullinfo()\fP function can be used to find out how many capturing
  subpatterns there are in a compiled pattern. The smallest size for
  \fIovector\fP that will allow for \fIn\fP captured substrings, in addition to
  the offsets of the substring matched by the whole pattern, is (\fIn\fP+1)*3.
+ .P
+ It is possible for capturing subpattern number \fIn+1\fP to match some part of
+ the subject when subpattern \fIn\fP has not been used at all. For example, if
+ the string "abc" is matched against the pattern (a|(z))(bc) the return from the
+ function is 4, and subpatterns 1 and 3 are matched, but 2 is not. When this
+ happens, both values in the offset pairs corresponding to unused subpatterns
+ are set to -1.
+ .P
+ Offset values that correspond to unused subpatterns at the end of the
+ expression are also set to -1. For example, if the string "abc" is matched
+ against the pattern (abc)(x(yz)?)? subpatterns 2 and 3 are not matched. The
+ return from the function is 2, because the highest used capturing subpattern
+ number is 1, and the offsets for for the second and third capturing subpatterns
+ (assuming the vector is large enough, of course) are set to -1.
+ .P
+ \fBNote\fP: Elements in the first two-thirds of \fIovector\fP that do not
+ correspond to capturing parentheses in the pattern are never changed. That is,
+ if a pattern contains \fIn\fP capturing parentheses, no more than
+ \fIovector[0]\fP to \fIovector[2n+1]\fP are set by \fBpcre_exec()\fP. The other
+ elements (in the first two-thirds) retain whatever values they previously had.
+ .P
+ Some convenience functions are provided for extracting the captured substrings
+ as separate strings. These are described below.
+ .
  .
- .SS "Return values from \fBpcre_exec()\fP"
+ .\" HTML <a name="errorlist"></a>
+ .SS "Error return values from \fBpcre_exec()\fP"
  .rs
  .sp
  If \fBpcre_exec()\fP fails, it returns a negative number. The following are
-Line 1068 
 compiled in an environment of one endian
+Line 1888 
 compiled in an environment of one endian
  other endianness. This is the error that PCRE gives when the magic number is
  not present.
  .sp
-   PCRE_ERROR_UNKNOWN_NODE   (-5)
+   PCRE_ERROR_UNKNOWN_OPCODE (-5)
  .sp
  While running the pattern match, an unknown item was encountered in the
  compiled pattern. This error could be caused by a bug in PCRE or by overwriting
-Line 1081 
 If a pattern contains back references, b
+Line 1901 
 If a pattern contains back references, b
  gets a block of memory at the start of matching to use for this purpose. If the
  call via \fBpcre_malloc()\fP fails, this error is given. The memory is
  automatically freed at the end of matching.
+ .P
+ This error is also given if \fBpcre_stack_malloc()\fP fails in
+ \fBpcre_exec()\fP. This can happen only when PCRE has been compiled with
+ \fB--disable-stack-for-recursion\fP.
  .sp
    PCRE_ERROR_NOSUBSTRING    (-7)
  .sp
-Line 1090 
 below). It is never returned by \fBpcre_
+Line 1914 
 below). It is never returned by \fBpcre_
  .sp
    PCRE_ERROR_MATCHLIMIT     (-8)
  .sp
- The recursion and backtracking limit, as specified by the \fImatch_limit\fP
+ The backtracking limit, as specified by the \fImatch_limit\fP field in a
- field in a \fBpcre_extra\fP structure (or defaulted) was reached. See the
+ \fBpcre_extra\fP structure (or defaulted) was reached. See the description
- description above.
+ above.
  .sp
    PCRE_ERROR_CALLOUT        (-9)
  .sp
-Line 1105 
 documentation for details.
+Line 1929 
 documentation for details.
  .sp
    PCRE_ERROR_BADUTF8        (-10)
  .sp
- 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,
+ and the PCRE_NO_UTF8_CHECK option was not set. If the size of the output vector
+ (\fIovecsize\fP) is at least 2, the byte offset to the start of the the invalid
+ UTF-8 character is placed in the first element, and a reason code is placed in
+ the second element. The reason codes are listed in the
+ .\" HTML <a href="#badutf8reasons">
+ .\" </a>
+ following section.
+ .\"
+ For backward compatibility, if PCRE_PARTIAL_HARD is set and the problem is a
+ truncated UTF-8 character at the end of the subject (reason codes 1 to 5),
+ PCRE_ERROR_SHORTUTF8 is returned instead of PCRE_ERROR_BADUTF8.
  .sp
    PCRE_ERROR_BADUTF8_OFFSET (-11)
  .sp
- 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
- 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
+ \fIstartoffset\fP did not point to the beginning of a UTF-8 character or the
+ end of the subject.
  .sp
-   PCRE_ERROR_PARTIAL (-12)
+   PCRE_ERROR_PARTIAL        (-12)
  .sp
  The subject string did not match, but it did match partially. See the
  .\" HREF
-Line 1120 
 The subject string did not match, but it
+Line 1957 
 The subject string did not match, but it
  .\"
  documentation for details of partial matching.
  .sp
-   PCRE_ERROR_BAD_PARTIAL (-13)
+   PCRE_ERROR_BADPARTIAL     (-13)
  .sp
- The PCRE_PARTIAL option was used with a compiled pattern containing items that
+ This code is no longer in use. It was formerly returned when the PCRE_PARTIAL
- are not supported for partial matching. See the
+ option was used with a compiled pattern containing items that were not
- .\" HREF
+ supported for partial matching. From release 8.00 onwards, there are no
- \fBpcrepartial\fP
+ restrictions on partial matching.
- .\"
- documentation for details of partial matching.
  .sp
-   PCRE_ERROR_INTERNAL (-14)
+   PCRE_ERROR_INTERNAL       (-14)
  .sp
  An unexpected internal error has occurred. This error could be caused by a bug
  in PCRE or by overwriting of the compiled pattern.
  .sp
-   PCRE_ERROR_BADCOUNT (-15)
+   PCRE_ERROR_BADCOUNT       (-15)
  .sp
  This error is given if the value of the \fIovecsize\fP argument is negative.
+ .sp
+   PCRE_ERROR_RECURSIONLIMIT (-21)
+ .sp
+ The internal recursion limit, as specified by the \fImatch_limit_recursion\fP
+ field in a \fBpcre_extra\fP structure (or defaulted) was reached. See the
+ description above.
+ .sp
+   PCRE_ERROR_BADNEWLINE     (-23)
+ .sp
+ An invalid combination of PCRE_NEWLINE_\fIxxx\fP options was given.
+ .sp
+   PCRE_ERROR_BADOFFSET      (-24)
+ .sp
+ The value of \fIstartoffset\fP was negative or greater than the length of the
+ subject, that is, the value in \fIlength\fP.
+ .sp
+   PCRE_ERROR_SHORTUTF8      (-25)
+ .sp
+ This error is returned instead of PCRE_ERROR_BADUTF8 when the subject string
+ ends with a truncated UTF-8 character and the PCRE_PARTIAL_HARD option is set.
+ Information about the failure is returned as for PCRE_ERROR_BADUTF8. It is in
+ fact sufficient to detect this case, but this special error code for
+ PCRE_PARTIAL_HARD precedes the implementation of returned information; it is
+ retained for backwards compatibility.
+ .sp
+   PCRE_ERROR_RECURSELOOP    (-26)
+ .sp
+ This error is returned when \fBpcre_exec()\fP detects a recursion loop within
+ the pattern. Specifically, it means that either the whole pattern or a
+ subpattern has been called recursively for the second time at the same position
+ in the subject string. Some simple patterns that might do this are detected and
+ faulted at compile time, but more complicated cases, in particular mutual
+ recursions between two different subpatterns, cannot be detected until run
+ time.
+ .sp
+   PCRE_ERROR_JIT_STACKLIMIT (-27)
+ .sp
+ This error is returned when a pattern that was successfully studied using the
+ PCRE_STUDY_JIT_COMPILE option is being matched, but the memory available for
+ the just-in-time processing stack is not large enough. See the
+ .\" HREF
+ \fBpcrejit\fP
+ .\"
+ documentation for more details.
+ .P
+ Error numbers -16 to -20 and -22 are not used by \fBpcre_exec()\fP.
+ .
+ .
+ .\" HTML <a name="badutf8reasons"></a>
+ .SS "Reason codes for invalid UTF-8 strings"
+ .rs
+ .sp
+ When \fBpcre_exec()\fP returns either PCRE_ERROR_BADUTF8 or
+ PCRE_ERROR_SHORTUTF8, and the size of the output vector (\fIovecsize\fP) is at
+ least 2, the offset of the start of the invalid UTF-8 character is placed in
+ the first output vector element (\fIovector[0]\fP) and a reason code is placed
+ in the second element (\fIovector[1]\fP). The reason codes are given names in
+ the \fBpcre.h\fP header file:
+ .sp
+   PCRE_UTF8_ERR1
+   PCRE_UTF8_ERR2
+   PCRE_UTF8_ERR3
+   PCRE_UTF8_ERR4
+   PCRE_UTF8_ERR5
+ .sp
+ The string ends with a truncated UTF-8 character; the code specifies how many
+ bytes are missing (1 to 5). Although RFC 3629 restricts UTF-8 characters to be
+ no longer than 4 bytes, the encoding scheme (originally defined by RFC 2279)
+ allows for up to 6 bytes, and this is checked first; hence the possibility of
+or 5 missing bytes.
+ .sp
+   PCRE_UTF8_ERR6
+   PCRE_UTF8_ERR7
+   PCRE_UTF8_ERR8
+   PCRE_UTF8_ERR9
+   PCRE_UTF8_ERR10
+ .sp
+ The two most significant bits of the 2nd, 3rd, 4th, 5th, or 6th byte of the
+ character do not have the binary value 0b10 (that is, either the most
+ significant bit is 0, or the next bit is 1).
+ .sp
+   PCRE_UTF8_ERR11
+   PCRE_UTF8_ERR12
+ .sp
+ A character that is valid by the RFC 2279 rules is either 5 or 6 bytes long;
+ these code points are excluded by RFC 3629.
+ .sp
+   PCRE_UTF8_ERR13
+ .sp
+ A 4-byte character has a value greater than 0x10fff; these code points are
+ excluded by RFC 3629.
+ .sp
+   PCRE_UTF8_ERR14
+ .sp
+ A 3-byte character has a value in the range 0xd800 to 0xdfff; this range of
+ code points are reserved by RFC 3629 for use with UTF-16, and so are excluded
+ from UTF-8.
+ .sp
+   PCRE_UTF8_ERR15
+   PCRE_UTF8_ERR16
+   PCRE_UTF8_ERR17
+   PCRE_UTF8_ERR18
+   PCRE_UTF8_ERR19
+ .sp
+ A 2-, 3-, 4-, 5-, or 6-byte character is "overlong", that is, it codes for a
+ value that can be represented by fewer bytes, which is invalid. For example,
+ the two bytes 0xc0, 0xae give the value 0x2e, whose correct coding uses just
+ one byte.
+ .sp
+   PCRE_UTF8_ERR20
+ .sp
+ The two most significant bits of the first byte of a character have the binary
+ value 0b10 (that is, the most significant bit is 1 and the second is 0). Such a
+ byte can only validly occur as the second or subsequent byte of a multi-byte
+ character.
+ .sp
+   PCRE_UTF8_ERR21
+ .sp
+ The first byte of a character has the value 0xfe or 0xff. These values can
+ never occur in a valid UTF-8 string.
  .
  .
  .SH "EXTRACTING CAPTURED SUBSTRINGS BY NUMBER"
-Line 1148 
 This error is given if the value of the
+Line 2103 
 This error is given if the value of the
  .ti +5n
  .B int \fIbuffersize\fP);
  .PP
- .br
  .B int pcre_get_substring(const char *\fIsubject\fP, int *\fIovector\fP,
  .ti +5n
  .B int \fIstringcount\fP, int \fIstringnumber\fP,
  .ti +5n
  .B const char **\fIstringptr\fP);
  .PP
- .br
  .B int pcre_get_substring_list(const char *\fIsubject\fP,
  .ti +5n
  .B int *\fIovector\fP, int \fIstringcount\fP, "const char ***\fIlistptr\fP);"
-Line 1166 
 Captured substrings can be accessed dire
+Line 2119 
 Captured substrings can be accessed dire
  \fBpcre_get_substring_list()\fP are provided for extracting captured substrings
  as new, separate, zero-terminated strings. These functions identify substrings
  by number. The next section describes functions for extracting named
- substrings. A substring that contains a binary zero is correctly extracted and
+ substrings.
- has a further zero added on the end, but the result is not, of course,
+ .P
- a C string.
+ A substring that contains a binary zero is correctly extracted and has a
+ further zero added on the end, but the result is not, of course, a C string.
+ However, you can process such a string by referring to the length that is
+ returned by \fBpcre_copy_substring()\fP and \fBpcre_get_substring()\fP.
+ Unfortunately, the interface to \fBpcre_get_substring_list()\fP is not adequate
+ for handling strings containing binary zeros, because the end of the final
+ string is not independently indicated.
  .P
  The first three arguments are the same for all three of these functions:
  \fIsubject\fP is the subject string that has just been successfully matched,
-Line 1188 
 the string is placed in \fIbuffer\fP, wh
+Line 2147 
 the string is placed in \fIbuffer\fP, wh
  \fIbuffersize\fP, while for \fBpcre_get_substring()\fP a new block of memory is
  obtained via \fBpcre_malloc\fP, and its address is returned via
  \fIstringptr\fP. The yield of the function is the length of the string, not
- including the terminating zero, or one of
+ including the terminating zero, or one of these error codes:
  .sp
    PCRE_ERROR_NOMEMORY       (-6)
  .sp
-Line 1204 
 and builds a list of pointers to them. A
+Line 2163 
 and builds a list of pointers to them. A
  memory that is obtained via \fBpcre_malloc\fP. The address of the memory block
  is returned via \fIlistptr\fP, which is also the start of the list of string
  pointers. The end of the list is marked by a NULL pointer. The yield of the
- function is zero if all went well, or
+ function is zero if all went well, or the error code
  .sp
    PCRE_ERROR_NOMEMORY       (-6)
  .sp
-Line 1223 
 a previous call of \fBpcre_get_substring
+Line 2182 
 a previous call of \fBpcre_get_substring
  \fBpcre_get_substring_list()\fP, respectively. They do nothing more than call
  the function pointed to by \fBpcre_free\fP, which of course could be called
  directly from a C program. However, PCRE is used in some situations where it is
- linked via a special interface to another programming language which cannot use
+ linked via a special interface to another programming language that cannot use
  \fBpcre_free\fP directly; it is for these cases that the functions are
  provided.
  .
-Line 1235 
 provided.
+Line 2194 
 provided.
  .ti +5n
  .B const char *\fIname\fP);
  .PP
- .br
  .B int pcre_copy_named_substring(const pcre *\fIcode\fP,
  .ti +5n
  .B const char *\fIsubject\fP, int *\fIovector\fP,
-Line 1244 
 provided.
+Line 2202 
 provided.
  .ti +5n
  .B char *\fIbuffer\fP, int \fIbuffersize\fP);
  .PP
- .br
  .B int pcre_get_named_substring(const pcre *\fIcode\fP,
  .ti +5n
  .B const char *\fIsubject\fP, int *\fIovector\fP,
-Line 1258 
 For example, for this pattern
+Line 2215 
 For example, for this pattern
  .sp
    (a+)b(?<xxx>\ed+)...
  .sp
- the number of the subpattern called "xxx" is 2. You can find the number from
+ the number of the subpattern called "xxx" is 2. If the name is known to be
- the name by calling \fBpcre_get_stringnumber()\fP. The first argument is the
+ unique (PCRE_DUPNAMES was not set), you can find the number from the name by
- compiled pattern, and the second is the name. The yield of the function is the
+ calling \fBpcre_get_stringnumber()\fP. The first argument is the compiled
+ pattern, and the second is the name. The yield of the function is the
  subpattern number, or PCRE_ERROR_NOSUBSTRING (-7) if there is no subpattern of
  that name.
  .P
-Line 1268 
 Given the number, you can extract the su
+Line 2226 
 Given the number, you can extract the su
  functions described in the previous section. For convenience, there are also
  two functions that do the whole job.
  .P
- Most of the arguments of \fIpcre_copy_named_substring()\fP and
+ Most of the arguments of \fBpcre_copy_named_substring()\fP and
- \fIpcre_get_named_substring()\fP are the same as those for the similarly named
+ \fBpcre_get_named_substring()\fP are the same as those for the similarly named
  functions that extract by number. As these are described in the previous
  section, they are not re-described here. There are just two differences:
  .P
-Line 1279 
 pattern. This is needed in order to gain
+Line 2237 
 pattern. This is needed in order to gain
  translation table.
  .P
  These functions call \fBpcre_get_stringnumber()\fP, and if it succeeds, they
- then call \fIpcre_copy_substring()\fP or \fIpcre_get_substring()\fP, as
+ then call \fBpcre_copy_substring()\fP or \fBpcre_get_substring()\fP, as
- appropriate.
+ appropriate. \fBNOTE:\fP If PCRE_DUPNAMES is set and there are duplicate names,
+ the behaviour may not be what you want (see the next section).
+ .P
+ \fBWarning:\fP If the pattern uses the (?| feature to set up multiple
+ subpatterns with the same number, as described in the
+ .\" HTML <a href="pcrepattern.html#dupsubpatternnumber">
+ .\" </a>
+ section on duplicate subpattern numbers
+ .\"
+ in the
+ .\" HREF
+ \fBpcrepattern\fP
+ .\"
+ page, you cannot use names to distinguish the different subpatterns, because
+ names are not included in the compiled code. The matching process uses only
+ numbers. For this reason, the use of different names for subpatterns of the
+ same number causes an error at compile time.
+ .
+ .
+ .SH "DUPLICATE SUBPATTERN NAMES"
+ .rs
+ .sp
+ .B int pcre_get_stringtable_entries(const pcre *\fIcode\fP,
+ .ti +5n
+ .B const char *\fIname\fP, char **\fIfirst\fP, char **\fIlast\fP);
+ .PP
+ When a pattern is compiled with the PCRE_DUPNAMES option, names for subpatterns
+ are not required to be unique. (Duplicate names are always allowed for
+ subpatterns with the same number, created by using the (?| feature. Indeed, if
+ such subpatterns are named, they are required to use the same names.)
+ .P
+ Normally, patterns with duplicate names are such that in any one match, only
+ one of the named subpatterns participates. An example is shown in the
+ .\" HREF
+ \fBpcrepattern\fP
+ .\"
+ documentation.
+ .P
+ When duplicates are present, \fBpcre_copy_named_substring()\fP and
+ \fBpcre_get_named_substring()\fP return the first substring corresponding to
+ the given name that is set. If none are set, PCRE_ERROR_NOSUBSTRING (-7) is
+ returned; no data is returned. The \fBpcre_get_stringnumber()\fP function
+ returns one of the numbers that are associated with the name, but it is not
+ defined which it is.
+ .P
+ If you want to get full details of all captured substrings for a given name,
+ you must use the \fBpcre_get_stringtable_entries()\fP function. The first
+ argument is the compiled pattern, and the second is the name. The third and
+ fourth are pointers to variables which are updated by the function. After it
+ has run, they point to the first and last entries in the name-to-number table
+ for the given name. The function itself returns the length of each entry, or
+ PCRE_ERROR_NOSUBSTRING (-7) if there are none. The format of the table is
+ described above in the section entitled \fIInformation about a pattern\fP
+ .\" HTML <a href="#infoaboutpattern">
+ .\" </a>
+ above.
+ .\"
+ Given all the relevant entries for the name, you can extract each of their
+ numbers, and hence the captured data, if any.
+ .
+ .
+ .SH "FINDING ALL POSSIBLE MATCHES"
+ .rs
+ .sp
+ The traditional matching function uses a similar algorithm to Perl, which stops
+ when it finds the first match, starting at a given point in the subject. If you
+ want to find all possible matches, or the longest possible match, consider
+ using the alternative matching function (see below) instead. If you cannot use
+ the alternative function, but still need to find all possible matches, you
+ can kludge it up by making use of the callout facility, which is described in
+ the
+ .\" HREF
+ \fBpcrecallout\fP
+ .\"
+ documentation.
+ .P
+ What you have to do is to insert a callout right at the end of the pattern.
+ When your callout function is called, extract and save the current matched
+ substring. Then return 1, which forces \fBpcre_exec()\fP to backtrack and try
+ other alternatives. Ultimately, when it runs out of matches, \fBpcre_exec()\fP
+ will yield PCRE_ERROR_NOMATCH.
+ .
+ .
+ .\" HTML <a name="dfamatch"></a>
+ .SH "MATCHING A PATTERN: THE ALTERNATIVE FUNCTION"
+ .rs
+ .sp
+ .B int pcre_dfa_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
+ .ti +5n
+ .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
+ .ti +5n
+ .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP,
+ .ti +5n
+ .B int *\fIworkspace\fP, int \fIwscount\fP);
+ .P
+ The function \fBpcre_dfa_exec()\fP is called to match a subject string against
+ a compiled pattern, using a matching algorithm that scans the subject string
+ just once, and does not backtrack. This has different characteristics to the
+ normal algorithm, and is not compatible with Perl. Some of the features of PCRE
+ patterns are not supported. Nevertheless, there are times when this kind of
+ matching can be useful. For a discussion of the two matching algorithms, and a
+ list of features that \fBpcre_dfa_exec()\fP does not support, see the
+ .\" HREF
+ \fBpcrematching\fP
+ .\"
+ documentation.
+ .P
+ The arguments for the \fBpcre_dfa_exec()\fP function are the same as for
+ \fBpcre_exec()\fP, plus two extras. The \fIovector\fP argument is used in a
+ different way, and this is described below. The other common arguments are used
+ in the same way as for \fBpcre_exec()\fP, so their description is not repeated
+ here.
+ .P
+ The two additional arguments provide workspace for the function. The workspace
+ vector should contain at least 20 elements. It is used for keeping track of
+ multiple paths through the pattern tree. More workspace will be needed for
+ patterns and subjects where there are a lot of potential matches.
+ .P
+ Here is an example of a simple call to \fBpcre_dfa_exec()\fP:
+ .sp
+   int rc;
+   int ovector[10];
+   int wspace[20];
+   rc = pcre_dfa_exec(
+     re,             /* result of pcre_compile() */
+     NULL,           /* we didn't study the pattern */
+     "some string",  /* the subject string */
+,             /* the length of the subject string */
+,              /* start at offset 0 in the subject */
+,              /* default options */
+     ovector,        /* vector of integers for substring information */
+,             /* number of elements (NOT size in bytes) */
+     wspace,         /* working space vector */
+);            /* number of elements (NOT size in bytes) */
+ .
+ .SS "Option bits for \fBpcre_dfa_exec()\fP"
+ .rs
+ .sp
+ The unused bits of the \fIoptions\fP argument for \fBpcre_dfa_exec()\fP must be
+ zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_\fIxxx\fP,
+ PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NOTEMPTY_ATSTART,
+ PCRE_NO_UTF8_CHECK, PCRE_BSR_ANYCRLF, PCRE_BSR_UNICODE, PCRE_NO_START_OPTIMIZE,
+ PCRE_PARTIAL_HARD, PCRE_PARTIAL_SOFT, PCRE_DFA_SHORTEST, and PCRE_DFA_RESTART.
+ All but the last four of these are exactly the same as for \fBpcre_exec()\fP,
+ so their description is not repeated here.
+ .sp
+   PCRE_PARTIAL_HARD
+   PCRE_PARTIAL_SOFT
+ .sp
+ These have the same general effect as they do for \fBpcre_exec()\fP, but the
+ details are slightly different. When PCRE_PARTIAL_HARD is set for
+ \fBpcre_dfa_exec()\fP, it returns PCRE_ERROR_PARTIAL if the end of the subject
+ is reached and there is still at least one matching possibility that requires
+ additional characters. This happens even if some complete matches have also
+ been found. When PCRE_PARTIAL_SOFT is set, the return code PCRE_ERROR_NOMATCH
+ is converted into PCRE_ERROR_PARTIAL if the end of the subject is reached,
+ there have been no complete matches, but there is still at least one matching
+ possibility. The portion of the string that was inspected when the longest
+ partial match was found is set as the first matching string in both cases.
+ There is a more detailed discussion of partial and multi-segment matching, with
+ examples, in the
+ .\" HREF
+ \fBpcrepartial\fP
+ .\"
+ documentation.
+ .sp
+   PCRE_DFA_SHORTEST
+ .sp
+ Setting the PCRE_DFA_SHORTEST option causes the matching algorithm to stop as
+ soon as it has found one match. Because of the way the alternative algorithm
+ works, this is necessarily the shortest possible match at the first possible
+ matching point in the subject string.
+ .sp
+   PCRE_DFA_RESTART
+ .sp
+ When \fBpcre_dfa_exec()\fP returns a partial match, it is possible to call it
+ again, with additional subject characters, and have it continue with the same
+ match. The PCRE_DFA_RESTART option requests this action; when it is set, the
+ \fIworkspace\fP and \fIwscount\fP options must reference the same vector as
+ before because data about the match so far is left in them after a partial
+ match. There is more discussion of this facility in the
+ .\" HREF
+ \fBpcrepartial\fP
+ .\"
+ documentation.
+ .
+ .
+ .SS "Successful returns from \fBpcre_dfa_exec()\fP"
+ .rs
+ .sp
+ When \fBpcre_dfa_exec()\fP succeeds, it may have matched more than one
+ substring in the subject. Note, however, that all the matches from one run of
+ the function start at the same point in the subject. The shorter matches are
+ all initial substrings of the longer matches. For example, if the pattern
+ .sp
+   <.*>
+ .sp
+ is matched against the string
+ .sp
+   This is <something> <something else> <something further> no more
+ .sp
+ the three matched strings are
+ .sp
+   <something>
+   <something> <something else>
+   <something> <something else> <something further>
+ .sp
+ On success, the yield of the function is a number greater than zero, which is
+ the number of matched substrings. The substrings themselves are returned in
+ \fIovector\fP. Each string uses two elements; the first is the offset to the
+ start, and the second is the offset to the end. In fact, all the strings have
+ the same start offset. (Space could have been saved by giving this only once,
+ but it was decided to retain some compatibility with the way \fBpcre_exec()\fP
+ returns data, even though the meaning of the strings is different.)
  .P
- .in 0
+ The strings are returned in reverse order of length; that is, the longest
- Last updated: 09 September 2004
+ matching string is given first. If there were too many matches to fit into
- .br
+ \fIovector\fP, the yield of the function is zero, and the vector is filled with
- Copyright (c) 1997-2004 University of Cambridge.
+ the longest matches. Unlike \fBpcre_exec()\fP, \fBpcre_dfa_exec()\fP can use
+ the entire \fIovector\fP for returning matched strings.
+ .
+ .
+ .SS "Error returns from \fBpcre_dfa_exec()\fP"
+ .rs
+ .sp
+ The \fBpcre_dfa_exec()\fP function returns a negative number when it fails.
+ Many of the errors are the same as for \fBpcre_exec()\fP, and these are
+ described
+ .\" HTML <a href="#errorlist">
+ .\" </a>
+ above.
+ .\"
+ There are in addition the following errors that are specific to
+ \fBpcre_dfa_exec()\fP:
+ .sp
+   PCRE_ERROR_DFA_UITEM      (-16)
+ .sp
+ This return is given if \fBpcre_dfa_exec()\fP encounters an item in the pattern
+ that it does not support, for instance, the use of \eC or a back reference.
+ .sp
+   PCRE_ERROR_DFA_UCOND      (-17)
+ .sp
+ This return is given if \fBpcre_dfa_exec()\fP encounters a condition item that
+ uses a back reference for the condition, or a test for recursion in a specific
+ group. These are not supported.
+ .sp
+   PCRE_ERROR_DFA_UMLIMIT    (-18)
+ .sp
+ This return is given if \fBpcre_dfa_exec()\fP is called with an \fIextra\fP
+ block that contains a setting of the \fImatch_limit\fP or
+ \fImatch_limit_recursion\fP fields. This is not supported (these fields are
+ meaningless for DFA matching).
+ .sp
+   PCRE_ERROR_DFA_WSSIZE     (-19)
+ .sp
+ This return is given if \fBpcre_dfa_exec()\fP runs out of space in the
+ \fIworkspace\fP vector.
+ .sp
+   PCRE_ERROR_DFA_RECURSE    (-20)
+ .sp
+ When a recursive subpattern is processed, the matching function calls itself
+ recursively, using private vectors for \fIovector\fP and \fIworkspace\fP. This
+ error is given if the output vector is not large enough. This should be
+ extremely rare, as a vector of size 1000 is used.
+ .
+ .
+ .SH "SEE ALSO"
+ .rs
+ .sp
+ \fBpcrebuild\fP(3), \fBpcrecallout\fP(3), \fBpcrecpp(3)\fP(3),
+ \fBpcrematching\fP(3), \fBpcrepartial\fP(3), \fBpcreposix\fP(3),
+ \fBpcreprecompile\fP(3), \fBpcresample\fP(3), \fBpcrestack\fP(3).
+ .
+ .
+ .SH AUTHOR
+ .rs
+ .sp
+ .nf
+ Philip Hazel
+ University Computing Service
+ Cambridge CB2 3QH, England.
+ .fi
+ .
+ .
+ .SH REVISION
+ .rs
+ .sp
+ .nf
+ Last updated: 06 September 2011
+ Copyright (c) 1997-2011 University of Cambridge.
+ .fi

 Legend:



Removed from v.75
 


changed lines


 
Added in v.686
 Legend:



Removed from v.75
 


changed lines


 
Added in v.686
-Removed from v.75
+Added in v.686

webmaster@exim.org	ViewVC Help
Powered by ViewVC 1.1.12