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

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

Parent Directory Parent Directory | Revision Log Revision Log | View Patch Patch

revision 568 by ph10, Sat Nov 6 17:36:26 2010 UTC revision 978 by ph10, Sun Jun 17 16:55:07 2012 UTC
# Line 1  Line 1 
1  .TH PCREAPI 3  .TH PCREAPI 3 "04 May 2012" "PCRE 8.31"
2  .SH NAME  .SH NAME
3  PCRE - Perl-compatible regular expressions  PCRE - Perl-compatible regular expressions
 .SH "PCRE NATIVE API"  
 .rs  
4  .sp  .sp
5  .B #include <pcre.h>  .B #include <pcre.h>
6  .PP  .
7    .
8    .SH "PCRE NATIVE API BASIC FUNCTIONS"
9    .rs
10    .sp
11  .SM  .SM
12  .B pcre *pcre_compile(const char *\fIpattern\fP, int \fIoptions\fP,  .B pcre *pcre_compile(const char *\fIpattern\fP, int \fIoptions\fP,
13  .ti +5n  .ti +5n
# Line 25  PCRE - Perl-compatible regular expressio Line 27  PCRE - Perl-compatible regular expressio
27  .ti +5n  .ti +5n
28  .B const char **\fIerrptr\fP);  .B const char **\fIerrptr\fP);
29  .PP  .PP
30    .B void pcre_free_study(pcre_extra *\fIextra\fP);
31    .PP
32  .B int pcre_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"  .B int pcre_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
33  .ti +5n  .ti +5n
34  .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,  .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
# Line 38  PCRE - Perl-compatible regular expressio Line 42  PCRE - Perl-compatible regular expressio
42  .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP,  .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP,
43  .ti +5n  .ti +5n
44  .B int *\fIworkspace\fP, int \fIwscount\fP);  .B int *\fIworkspace\fP, int \fIwscount\fP);
45  .PP  .
46    .
47    .SH "PCRE NATIVE API STRING EXTRACTION FUNCTIONS"
48    .rs
49    .sp
50  .B int pcre_copy_named_substring(const pcre *\fIcode\fP,  .B int pcre_copy_named_substring(const pcre *\fIcode\fP,
51  .ti +5n  .ti +5n
52  .B const char *\fIsubject\fP, int *\fIovector\fP,  .B const char *\fIsubject\fP, int *\fIovector\fP,
# Line 82  PCRE - Perl-compatible regular expressio Line 90  PCRE - Perl-compatible regular expressio
90  .B void pcre_free_substring(const char *\fIstringptr\fP);  .B void pcre_free_substring(const char *\fIstringptr\fP);
91  .PP  .PP
92  .B void pcre_free_substring_list(const char **\fIstringptr\fP);  .B void pcre_free_substring_list(const char **\fIstringptr\fP);
93    .
94    .
95    .SH "PCRE NATIVE API AUXILIARY FUNCTIONS"
96    .rs
97    .sp
98    .B pcre_jit_stack *pcre_jit_stack_alloc(int \fIstartsize\fP, int \fImaxsize\fP);
99    .PP
100    .B void pcre_jit_stack_free(pcre_jit_stack *\fIstack\fP);
101    .PP
102    .B void pcre_assign_jit_stack(pcre_extra *\fIextra\fP,
103    .ti +5n
104    .B pcre_jit_callback \fIcallback\fP, void *\fIdata\fP);
105  .PP  .PP
106  .B const unsigned char *pcre_maketables(void);  .B const unsigned char *pcre_maketables(void);
107  .PP  .PP
# Line 89  PCRE - Perl-compatible regular expressio Line 109  PCRE - Perl-compatible regular expressio
109  .ti +5n  .ti +5n
110  .B int \fIwhat\fP, void *\fIwhere\fP);  .B int \fIwhat\fP, void *\fIwhere\fP);
111  .PP  .PP
 .B int pcre_info(const pcre *\fIcode\fP, int *\fIoptptr\fP, int  
 .B *\fIfirstcharptr\fP);  
 .PP  
112  .B int pcre_refcount(pcre *\fIcode\fP, int \fIadjust\fP);  .B int pcre_refcount(pcre *\fIcode\fP, int \fIadjust\fP);
113  .PP  .PP
114  .B int pcre_config(int \fIwhat\fP, void *\fIwhere\fP);  .B int pcre_config(int \fIwhat\fP, void *\fIwhere\fP);
115  .PP  .PP
116  .B char *pcre_version(void);  .B const char *pcre_version(void);
117  .PP  .PP
118    .B int pcre_pattern_to_host_byte_order(pcre *\fIcode\fP,
119    .ti +5n
120    .B pcre_extra *\fIextra\fP, const unsigned char *\fItables\fP);
121    .
122    .
123    .SH "PCRE NATIVE API INDIRECTED FUNCTIONS"
124    .rs
125    .sp
126  .B void *(*pcre_malloc)(size_t);  .B void *(*pcre_malloc)(size_t);
127  .PP  .PP
128  .B void (*pcre_free)(void *);  .B void (*pcre_free)(void *);
# Line 109  PCRE - Perl-compatible regular expressio Line 134  PCRE - Perl-compatible regular expressio
134  .B int (*pcre_callout)(pcre_callout_block *);  .B int (*pcre_callout)(pcre_callout_block *);
135  .  .
136  .  .
137    .SH "PCRE 8-BIT AND 16-BIT LIBRARIES"
138    .rs
139    .sp
140    From release 8.30, PCRE can be compiled as a library for handling 16-bit
141    character strings as well as, or instead of, the original library that handles
142    8-bit character strings. To avoid too much complication, this document
143    describes the 8-bit versions of the functions, with only occasional references
144    to the 16-bit library.
145    .P
146    The 16-bit functions operate in the same way as their 8-bit counterparts; they
147    just use different data types for their arguments and results, and their names
148    start with \fBpcre16_\fP instead of \fBpcre_\fP. For every option that has UTF8
149    in its name (for example, PCRE_UTF8), there is a corresponding 16-bit name with
150    UTF8 replaced by UTF16. This facility is in fact just cosmetic; the 16-bit
151    option names define the same bit values.
152    .P
153    References to bytes and UTF-8 in this document should be read as references to
154    16-bit data quantities and UTF-16 when using the 16-bit library, unless
155    specified otherwise. More details of the specific differences for the 16-bit
156    library are given in the
157    .\" HREF
158    \fBpcre16\fP
159    .\"
160    page.
161    .
162    .
163  .SH "PCRE API OVERVIEW"  .SH "PCRE API OVERVIEW"
164  .rs  .rs
165  .sp  .sp
166  PCRE has its own native API, which is described in this document. There are  PCRE has its own native API, which is described in this document. There are
167  also some wrapper functions that correspond to the POSIX regular expression  also some wrapper functions (for the 8-bit library only) that correspond to the
168  API. These are described in the  POSIX regular expression API, but they do not give access to all the
169    functionality. They are described in the
170  .\" HREF  .\" HREF
171  \fBpcreposix\fP  \fBpcreposix\fP
172  .\"  .\"
173  documentation. Both of these APIs define a set of C function calls. A C++  documentation. Both of these APIs define a set of C function calls. A C++
174  wrapper is distributed with PCRE. It is documented in the  wrapper (again for the 8-bit library only) is also distributed with PCRE. It is
175    documented in the
176  .\" HREF  .\" HREF
177  \fBpcrecpp\fP  \fBpcrecpp\fP
178  .\"  .\"
179  page.  page.
180  .P  .P
181  The native API C function prototypes are defined in the header file  The native API C function prototypes are defined in the header file
182  \fBpcre.h\fP, and on Unix systems the library itself is called \fBlibpcre\fP.  \fBpcre.h\fP, and on Unix-like systems the (8-bit) library itself is called
183  It can normally be accessed by adding \fB-lpcre\fP to the command for linking  \fBlibpcre\fP. It can normally be accessed by adding \fB-lpcre\fP to the
184  an application that uses PCRE. The header file defines the macros PCRE_MAJOR  command for linking an application that uses PCRE. The header file defines the
185  and PCRE_MINOR to contain the major and minor release numbers for the library.  macros PCRE_MAJOR and PCRE_MINOR to contain the major and minor release numbers
186  Applications can use these to include support for different releases of PCRE.  for the library. Applications can use these to include support for different
187    releases of PCRE.
188  .P  .P
189  In a Windows environment, if you want to statically link an application program  In a Windows environment, if you want to statically link an application program
190  against a non-dll \fBpcre.a\fP file, you must define PCRE_STATIC before  against a non-dll \fBpcre.a\fP file, you must define PCRE_STATIC before
# Line 152  documentation, and the Line 206  documentation, and the
206  .\"  .\"
207  documentation describes how to compile and run it.  documentation describes how to compile and run it.
208  .P  .P
209    Just-in-time compiler support is an optional feature of PCRE that can be built
210    in appropriate hardware environments. It greatly speeds up the matching
211    performance of many patterns. Simple programs can easily request that it be
212    used if available, by setting an option that is ignored when it is not
213    relevant. More complicated programs might need to make use of the functions
214    \fBpcre_jit_stack_alloc()\fP, \fBpcre_jit_stack_free()\fP, and
215    \fBpcre_assign_jit_stack()\fP in order to control the JIT code's memory usage.
216    These functions are discussed in the
217    .\" HREF
218    \fBpcrejit\fP
219    .\"
220    documentation.
221    .P
222  A second matching function, \fBpcre_dfa_exec()\fP, which is not  A second matching function, \fBpcre_dfa_exec()\fP, which is not
223  Perl-compatible, is also provided. This uses a different algorithm for the  Perl-compatible, is also provided. This uses a different algorithm for the
224  matching. The alternative algorithm finds all possible matches (at a given  matching. The alternative algorithm finds all possible matches (at a given
# Line 186  specialist use. Most commonly, no specia Line 253  specialist use. Most commonly, no specia
253  internal tables that are generated when PCRE is built are used.  internal tables that are generated when PCRE is built are used.
254  .P  .P
255  The function \fBpcre_fullinfo()\fP is used to find out information about a  The function \fBpcre_fullinfo()\fP is used to find out information about a
256  compiled pattern; \fBpcre_info()\fP is an obsolete version that returns only  compiled pattern. The function \fBpcre_version()\fP returns a pointer to a
257  some of the available information, but is retained for backwards compatibility.  string containing the version of PCRE and its date of release.
 The function \fBpcre_version()\fP returns a pointer to a string containing the  
 version of PCRE and its date of release.  
258  .P  .P
259  The function \fBpcre_refcount()\fP maintains a reference count in a data block  The function \fBpcre_refcount()\fP maintains a reference count in a data block
260  containing a compiled pattern. This is provided for the benefit of  containing a compiled pattern. This is provided for the benefit of
# Line 237  PCRE supports five different conventions Line 302  PCRE supports five different conventions
302  strings: a single CR (carriage return) character, a single LF (linefeed)  strings: a single CR (carriage return) character, a single LF (linefeed)
303  character, the two-character sequence CRLF, any of the three preceding, or any  character, the two-character sequence CRLF, any of the three preceding, or any
304  Unicode newline sequence. The Unicode newline sequences are the three just  Unicode newline sequence. The Unicode newline sequences are the three just
305  mentioned, plus the single characters VT (vertical tab, U+000B), FF (formfeed,  mentioned, plus the single characters VT (vertical tab, U+000B), FF (form feed,
306  U+000C), NEL (next line, U+0085), LS (line separator, U+2028), and PS  U+000C), NEL (next line, U+0085), LS (line separator, U+2028), and PS
307  (paragraph separator, U+2029).  (paragraph separator, U+2029).
308  .P  .P
# Line 282  callout function pointed to by \fBpcre_c Line 347  callout function pointed to by \fBpcre_c
347  .P  .P
348  The compiled form of a regular expression is not altered during matching, so  The compiled form of a regular expression is not altered during matching, so
349  the same compiled pattern can safely be used by several threads at once.  the same compiled pattern can safely be used by several threads at once.
350    .P
351    If the just-in-time optimization feature is being used, it needs separate
352    memory stack areas for each thread. See the
353    .\" HREF
354    \fBpcrejit\fP
355    .\"
356    documentation for more details.
357  .  .
358  .  .
359  .SH "SAVING PRECOMPILED PATTERNS FOR LATER USE"  .SH "SAVING PRECOMPILED PATTERNS FOR LATER USE"
# Line 293  which it was compiled. Details are given Line 365  which it was compiled. Details are given
365  .\" HREF  .\" HREF
366  \fBpcreprecompile\fP  \fBpcreprecompile\fP
367  .\"  .\"
368  documentation. However, compiling a regular expression with one version of PCRE  documentation, which includes a description of the
369  for use with a different version is not guaranteed to work and may cause  \fBpcre_pattern_to_host_byte_order()\fP function. However, compiling a regular
370  crashes.  expression with one version of PCRE for use with a different version is not
371    guaranteed to work and may cause crashes.
372  .  .
373  .  .
374  .SH "CHECKING BUILD-TIME OPTIONS"  .SH "CHECKING BUILD-TIME OPTIONS"
# Line 312  documentation has more details about the Line 385  documentation has more details about the
385  .P  .P
386  The first argument for \fBpcre_config()\fP is an integer, specifying which  The first argument for \fBpcre_config()\fP is an integer, specifying which
387  information is required; the second argument is a pointer to a variable into  information is required; the second argument is a pointer to a variable into
388  which the information is placed. The following information is available:  which the information is placed. The returned value is zero on success, or the
389    negative error code PCRE_ERROR_BADOPTION if the value in the first argument is
390    not recognized. The following information is available:
391  .sp  .sp
392    PCRE_CONFIG_UTF8    PCRE_CONFIG_UTF8
393  .sp  .sp
394  The output is an integer that is set to one if UTF-8 support is available;  The output is an integer that is set to one if UTF-8 support is available;
395  otherwise it is set to zero.  otherwise it is set to zero. If this option is given to the 16-bit version of
396    this function, \fBpcre16_config()\fP, the result is PCRE_ERROR_BADOPTION.
397    .sp
398      PCRE_CONFIG_UTF16
399    .sp
400    The output is an integer that is set to one if UTF-16 support is available;
401    otherwise it is set to zero. This value should normally be given to the 16-bit
402    version of this function, \fBpcre16_config()\fP. If it is given to the 8-bit
403    version of this function, the result is PCRE_ERROR_BADOPTION.
404  .sp  .sp
405    PCRE_CONFIG_UNICODE_PROPERTIES    PCRE_CONFIG_UNICODE_PROPERTIES
406  .sp  .sp
407  The output is an integer that is set to one if support for Unicode character  The output is an integer that is set to one if support for Unicode character
408  properties is available; otherwise it is set to zero.  properties is available; otherwise it is set to zero.
409  .sp  .sp
410      PCRE_CONFIG_JIT
411    .sp
412    The output is an integer that is set to one if support for just-in-time
413    compiling is available; otherwise it is set to zero.
414    .sp
415      PCRE_CONFIG_JITTARGET
416    .sp
417    The output is a pointer to a zero-terminated "const char *" string. If JIT
418    support is available, the string contains the name of the architecture for
419    which the JIT compiler is configured, for example "x86 32bit (little endian +
420    unaligned)". If JIT support is not available, the result is NULL.
421    .sp
422    PCRE_CONFIG_NEWLINE    PCRE_CONFIG_NEWLINE
423  .sp  .sp
424  The output is an integer whose value specifies the default character sequence  The output is an integer whose value specifies the default character sequence
# Line 343  or CRLF. The default can be overridden w Line 438  or CRLF. The default can be overridden w
438    PCRE_CONFIG_LINK_SIZE    PCRE_CONFIG_LINK_SIZE
439  .sp  .sp
440  The output is an integer that contains the number of bytes used for internal  The output is an integer that contains the number of bytes used for internal
441  linkage in compiled regular expressions. The value is 2, 3, or 4. Larger values  linkage in compiled regular expressions. For the 8-bit library, the value can
442  allow larger regular expressions to be compiled, at the expense of slower  be 2, 3, or 4. For the 16-bit library, the value is either 2 or 4 and is still
443  matching. The default value of 2 is sufficient for all but the most massive  a number of bytes. The default value of 2 is sufficient for all but the most
444  patterns, since it allows the compiled pattern to be up to 64K in size.  massive patterns, since it allows the compiled pattern to be up to 64K in size.
445    Larger values allow larger regular expressions to be compiled, at the expense
446    of slower matching.
447  .sp  .sp
448    PCRE_CONFIG_POSIX_MALLOC_THRESHOLD    PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
449  .sp  .sp
# Line 428  within the pattern (see the detailed des Line 525  within the pattern (see the detailed des
525  documentation). For those options that can be different in different parts of  documentation). For those options that can be different in different parts of
526  the pattern, the contents of the \fIoptions\fP argument specifies their  the pattern, the contents of the \fIoptions\fP argument specifies their
527  settings at the start of compilation and execution. The PCRE_ANCHORED,  settings at the start of compilation and execution. The PCRE_ANCHORED,
528  PCRE_BSR_\fIxxx\fP, and PCRE_NEWLINE_\fIxxx\fP options can be set at the time  PCRE_BSR_\fIxxx\fP, PCRE_NEWLINE_\fIxxx\fP, PCRE_NO_UTF8_CHECK, and
529  of matching as well as at compile time.  PCRE_NO_START_OPTIMIZE options can be set at the time of matching as well as at
530    compile time.
531  .P  .P
532  If \fIerrptr\fP is NULL, \fBpcre_compile()\fP returns NULL immediately.  If \fIerrptr\fP is NULL, \fBpcre_compile()\fP returns NULL immediately.
533  Otherwise, if compilation of a pattern fails, \fBpcre_compile()\fP returns  Otherwise, if compilation of a pattern fails, \fBpcre_compile()\fP returns
534  NULL, and sets the variable pointed to by \fIerrptr\fP to point to a textual  NULL, and sets the variable pointed to by \fIerrptr\fP to point to a textual
535  error message. This is a static string that is part of the library. You must  error message. This is a static string that is part of the library. You must
536  not try to free it. The byte offset from the start of the pattern to the  not try to free it. Normally, the offset from the start of the pattern to the
537  character that was being processed when the error was discovered is placed in  byte that was being processed when the error was discovered is placed in the
538  the variable pointed to by \fIerroffset\fP, which must not be NULL. If it is,  variable pointed to by \fIerroffset\fP, which must not be NULL (if it is, an
539  an immediate error is given. Some errors are not detected until checks are  immediate error is given). However, for an invalid UTF-8 string, the offset is
540  carried out when the whole pattern has been scanned; in this case the offset is  that of the first byte of the failing character.
541  set to the end of the pattern.  .P
542    Some errors are not detected until the whole pattern has been scanned; in these
543    cases, the offset passed back is the length of the pattern. Note that the
544    offset is in bytes, not characters, even in UTF-8 mode. It may sometimes point
545    into the middle of a UTF-8 character.
546  .P  .P
547  If \fBpcre_compile2()\fP is used instead of \fBpcre_compile()\fP, and the  If \fBpcre_compile2()\fP is used instead of \fBpcre_compile()\fP, and the
548  \fIerrorcodeptr\fP argument is not NULL, a non-zero error code number is  \fIerrorcodeptr\fP argument is not NULL, a non-zero error code number is
# Line 519  pattern. Line 621  pattern.
621  .sp  .sp
622    PCRE_DOTALL    PCRE_DOTALL
623  .sp  .sp
624  If this bit is set, a dot metacharater in the pattern matches all characters,  If this bit is set, a dot metacharacter in the pattern matches a character of
625  including those that indicate newline. Without it, a dot does not match when  any value, including one that indicates a newline. However, it only ever
626  the current position is at a newline. This option is equivalent to Perl's /s  matches one character, even if newlines are coded as CRLF. Without this option,
627  option, and it can be changed within a pattern by a (?s) option setting. A  a dot does not match when the current position is at a newline. This option is
628  negative class such as [^a] always matches newline characters, independent of  equivalent to Perl's /s option, and it can be changed within a pattern by a
629  the setting of this option.  (?s) option setting. A negative class such as [^a] always matches newline
630    characters, independent of the setting of this option.
631  .sp  .sp
632    PCRE_DUPNAMES    PCRE_DUPNAMES
633  .sp  .sp
# Line 539  documentation. Line 642  documentation.
642  .sp  .sp
643    PCRE_EXTENDED    PCRE_EXTENDED
644  .sp  .sp
645  If this bit is set, whitespace data characters in the pattern are totally  If this bit is set, white space data characters in the pattern are totally
646  ignored except when escaped or inside a character class. Whitespace does not  ignored except when escaped or inside a character class. White space does not
647  include the VT character (code 11). In addition, characters between an  include the VT character (code 11). In addition, characters between an
648  unescaped # outside a character class and the next newline, inclusive, are also  unescaped # outside a character class and the next newline, inclusive, are also
649  ignored. This is equivalent to Perl's /x option, and it can be changed within a  ignored. This is equivalent to Perl's /x option, and it can be changed within a
650  pattern by a (?x) option setting.  pattern by a (?x) option setting.
651  .P  .P
652    Which characters are interpreted as newlines is controlled by the options
653    passed to \fBpcre_compile()\fP or by a special sequence at the start of the
654    pattern, as described in the section entitled
655    .\" HTML <a href="pcrepattern.html#newlines">
656    .\" </a>
657    "Newline conventions"
658    .\"
659    in the \fBpcrepattern\fP documentation. Note that the end of this type of
660    comment is a literal newline sequence in the pattern; escape sequences that
661    happen to represent a newline do not count.
662    .P
663  This option makes it possible to include comments inside complicated patterns.  This option makes it possible to include comments inside complicated patterns.
664  Note, however, that this applies only to data characters. Whitespace characters  Note, however, that this applies only to data characters. White space characters
665  may never appear within special character sequences in a pattern, for example  may never appear within special character sequences in a pattern, for example
666  within the sequence (?( which introduces a conditional subpattern.  within the sequence (?( that introduces a conditional subpattern.
667  .sp  .sp
668    PCRE_EXTRA    PCRE_EXTRA
669  .sp  .sp
# Line 582  character). Thus, the pattern AB]CD beco Line 696  character). Thus, the pattern AB]CD beco
696  string (by default this causes the current matching alternative to fail). A  string (by default this causes the current matching alternative to fail). A
697  pattern such as (\e1)(a) succeeds when this option is set (assuming it can find  pattern such as (\e1)(a) succeeds when this option is set (assuming it can find
698  an "a" in the subject), whereas it fails by default, for Perl compatibility.  an "a" in the subject), whereas it fails by default, for Perl compatibility.
699    .P
700    (3) \eU matches an upper case "U" character; by default \eU causes a compile
701    time error (Perl uses \eU to upper case subsequent characters).
702    .P
703    (4) \eu matches a lower case "u" character unless it is followed by four
704    hexadecimal digits, in which case the hexadecimal number defines the code point
705    to match. By default, \eu causes a compile time error (Perl uses it to upper
706    case the following character).
707    .P
708    (5) \ex matches a lower case "x" character unless it is followed by two
709    hexadecimal digits, in which case the hexadecimal number defines the code point
710    to match. By default, as in Perl, a hexadecimal number is always expected after
711    \ex, but it may have zero, one, or two digits (so, for example, \exz matches a
712    binary zero character followed by z).
713  .sp  .sp
714    PCRE_MULTILINE    PCRE_MULTILINE
715  .sp  .sp
# Line 613  CRLF sequence. Setting PCRE_NEWLINE_ANYC Line 741  CRLF sequence. Setting PCRE_NEWLINE_ANYC
741  preceding sequences should be recognized. Setting PCRE_NEWLINE_ANY specifies  preceding sequences should be recognized. Setting PCRE_NEWLINE_ANY specifies
742  that any Unicode newline sequence should be recognized. The Unicode newline  that any Unicode newline sequence should be recognized. The Unicode newline
743  sequences are the three just mentioned, plus the single characters VT (vertical  sequences are the three just mentioned, plus the single characters VT (vertical
744  tab, U+000B), FF (formfeed, U+000C), NEL (next line, U+0085), LS (line  tab, U+000B), FF (form feed, U+000C), NEL (next line, U+0085), LS (line
745  separator, U+2028), and PS (paragraph separator, U+2029). The last two are  separator, U+2028), and PS (paragraph separator, U+2029). For the 8-bit
746  recognized only in UTF-8 mode.  library, the last two are recognized only in UTF-8 mode.
747  .P  .P
748  The newline setting in the options word uses three bits that are treated  The newline setting in the options word uses three bits that are treated
749  as a number, giving eight possibilities. Currently only six are used (default  as a number, giving eight possibilities. Currently only six are used (default
# Line 624  option, the combination may or may not b Line 752  option, the combination may or may not b
752  PCRE_NEWLINE_CR with PCRE_NEWLINE_LF is equivalent to PCRE_NEWLINE_CRLF, but  PCRE_NEWLINE_CR with PCRE_NEWLINE_LF is equivalent to PCRE_NEWLINE_CRLF, but
753  other combinations may yield unused numbers and cause an error.  other combinations may yield unused numbers and cause an error.
754  .P  .P
755  The only time that a line break is specially recognized when compiling a  The only time that a line break in a pattern is specially recognized when
756  pattern is if PCRE_EXTENDED is set, and an unescaped # outside a character  compiling is when PCRE_EXTENDED is set. CR and LF are white space characters,
757  class is encountered. This indicates a comment that lasts until after the next  and so are ignored in this mode. Also, an unescaped # outside a character class
758  line break sequence. In other circumstances, line break sequences are treated  indicates a comment that lasts until after the next line break sequence. In
759  as literal data, except that in PCRE_EXTENDED mode, both CR and LF are treated  other circumstances, line break sequences in patterns are treated as literal
760  as whitespace characters and are therefore ignored.  data.
761  .P  .P
762  The newline option that is set at compile time becomes the default that is used  The newline option that is set at compile time becomes the default that is used
763  for \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP, but it can be overridden.  for \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP, but it can be overridden.
# Line 642  were followed by ?: but named parenthese Line 770  were followed by ?: but named parenthese
770  they acquire numbers in the usual way). There is no equivalent of this option  they acquire numbers in the usual way). There is no equivalent of this option
771  in Perl.  in Perl.
772  .sp  .sp
773      NO_START_OPTIMIZE
774    .sp
775    This is an option that acts at matching time; that is, it is really an option
776    for \fBpcre_exec()\fP or \fBpcre_dfa_exec()\fP. If it is set at compile time,
777    it is remembered with the compiled pattern and assumed at matching time. For
778    details see the discussion of PCRE_NO_START_OPTIMIZE
779    .\" HTML <a href="#execoptions">
780    .\" </a>
781    below.
782    .\"
783    .sp
784    PCRE_UCP    PCRE_UCP
785  .sp  .sp
786  This option changes the way PCRE processes \eb, \ed, \es, \ew, and some of the  This option changes the way PCRE processes \eB, \eb, \eD, \ed, \eS, \es, \eW,
787  POSIX character classes. By default, only ASCII characters are recognized, but  \ew, and some of the POSIX character classes. By default, only ASCII characters
788  if PCRE_UCP is set, Unicode properties are used instead to classify characters.  are recognized, but if PCRE_UCP is set, Unicode properties are used instead to
789  More details are given in the section on  classify characters. More details are given in the section on
790  .\" HTML <a href="pcre.html#genericchartypes">  .\" HTML <a href="pcre.html#genericchartypes">
791  .\" </a>  .\" </a>
792  generic character types  generic character types
# Line 669  with Perl. It can also be set by a (?U) Line 808  with Perl. It can also be set by a (?U)
808    PCRE_UTF8    PCRE_UTF8
809  .sp  .sp
810  This option causes PCRE to regard both the pattern and the subject as strings  This option causes PCRE to regard both the pattern and the subject as strings
811  of UTF-8 characters instead of single-byte character strings. However, it is  of UTF-8 characters instead of single-byte strings. However, it is available
812  available only when PCRE is built to include UTF-8 support. If not, the use  only when PCRE is built to include UTF support. If not, the use of this option
813  of this option provokes an error. Details of how this option changes the  provokes an error. Details of how this option changes the behaviour of PCRE are
814  behaviour of PCRE are given in the  given in the
 .\" HTML <a href="pcre.html#utf8support">  
 .\" </a>  
 section on UTF-8 support  
 .\"  
 in the main  
815  .\" HREF  .\" HREF
816  \fBpcre\fP  \fBpcreunicode\fP
817  .\"  .\"
818  page.  page.
819  .sp  .sp
820    PCRE_NO_UTF8_CHECK    PCRE_NO_UTF8_CHECK
821  .sp  .sp
822  When PCRE_UTF8 is set, the validity of the pattern as a UTF-8 string is  When PCRE_UTF8 is set, the validity of the pattern as a UTF-8
823  automatically checked. There is a discussion about the  string is automatically checked. There is a discussion about the
824  .\" HTML <a href="pcre.html#utf8strings">  .\" HTML <a href="pcreunicode.html#utf8strings">
825  .\" </a>  .\" </a>
826  validity of UTF-8 strings  validity of UTF-8 strings
827  .\"  .\"
828  in the main  in the
829  .\" HREF  .\" HREF
830  \fBpcre\fP  \fBpcreunicode\fP
831  .\"  .\"
832  page. If an invalid UTF-8 sequence of bytes is found, \fBpcre_compile()\fP  page. If an invalid UTF-8 sequence is found, \fBpcre_compile()\fP returns an
833  returns an error. If you already know that your pattern is valid, and you want  error. If you already know that your pattern is valid, and you want to skip
834  to skip this check for performance reasons, you can set the PCRE_NO_UTF8_CHECK  this check for performance reasons, you can set the PCRE_NO_UTF8_CHECK option.
835  option. When it is set, the effect of passing an invalid UTF-8 string as a  When it is set, the effect of passing an invalid UTF-8 string as a pattern is
836  pattern is undefined. It may cause your program to crash. Note that this option  undefined. It may cause your program to crash. Note that this option can also
837  can also be passed to \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP, to suppress  be passed to \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP, to suppress the
838  the UTF-8 validity checking of subject strings.  validity checking of subject strings.
839  .  .
840  .  .
841  .SH "COMPILATION ERROR CODES"  .SH "COMPILATION ERROR CODES"
# Line 709  the UTF-8 validity checking of subject s Line 843  the UTF-8 validity checking of subject s
843  .sp  .sp
844  The following table lists the error codes than may be returned by  The following table lists the error codes than may be returned by
845  \fBpcre_compile2()\fP, along with the error messages that may be returned by  \fBpcre_compile2()\fP, along with the error messages that may be returned by
846  both compiling functions. As PCRE has developed, some error codes have fallen  both compiling functions. Note that error messages are always 8-bit ASCII
847  out of use. To avoid confusion, they have not been re-used.  strings, even in 16-bit mode. As PCRE has developed, some error codes have
848    fallen out of use. To avoid confusion, they have not been re-used.
849  .sp  .sp
850     0  no error     0  no error
851     1  \e at end of pattern     1  \e at end of pattern
# Line 744  out of use. To avoid confusion, they hav Line 879  out of use. To avoid confusion, they hav
879    29  (?R or (?[+-]digits must be followed by )    29  (?R or (?[+-]digits must be followed by )
880    30  unknown POSIX class name    30  unknown POSIX class name
881    31  POSIX collating elements are not supported    31  POSIX collating elements are not supported
882    32  this version of PCRE is not compiled with PCRE_UTF8 support    32  this version of PCRE is compiled without UTF support
883    33  [this code is not in use]    33  [this code is not in use]
884    34  character value in \ex{...} sequence is too large    34  character value in \ex{...} sequence is too large
885    35  invalid condition (?(0)    35  invalid condition (?(0)
886    36  \eC not allowed in lookbehind assertion    36  \eC not allowed in lookbehind assertion
887    37  PCRE does not support \eL, \el, \eN, \eU, or \eu    37  PCRE does not support \eL, \el, \eN{name}, \eU, or \eu
888    38  number after (?C is > 255    38  number after (?C is > 255
889    39  closing ) for (?C expected    39  closing ) for (?C expected
890    40  recursive call could loop indefinitely    40  recursive call could loop indefinitely
891    41  unrecognized character after (?P    41  unrecognized character after (?P
892    42  syntax error in subpattern name (missing terminator)    42  syntax error in subpattern name (missing terminator)
893    43  two named subpatterns have the same name    43  two named subpatterns have the same name
894    44  invalid UTF-8 string    44  invalid UTF-8 string (specifically UTF-8)
895    45  support for \eP, \ep, and \eX has not been compiled    45  support for \eP, \ep, and \eX has not been compiled
896    46  malformed \eP or \ep sequence    46  malformed \eP or \ep sequence
897    47  unknown property name after \eP or \ep    47  unknown property name after \eP or \ep
898    48  subpattern name is too long (maximum 32 characters)    48  subpattern name is too long (maximum 32 characters)
899    49  too many named subpatterns (maximum 10000)    49  too many named subpatterns (maximum 10000)
900    50  [this code is not in use]    50  [this code is not in use]
901    51  octal value is greater than \e377 (not in UTF-8 mode)    51  octal value is greater than \e377 in 8-bit non-UTF-8 mode
902    52  internal error: overran compiling workspace    52  internal error: overran compiling workspace
903    53  internal error: previously-checked referenced subpattern    53  internal error: previously-checked referenced subpattern
904          not found          not found
# Line 782  out of use. To avoid confusion, they hav Line 917  out of use. To avoid confusion, they hav
917    65  different names for subpatterns of the same number are    65  different names for subpatterns of the same number are
918          not allowed          not allowed
919    66  (*MARK) must have an argument    66  (*MARK) must have an argument
920    67  this version of PCRE is not compiled with PCRE_UCP support    67  this version of PCRE is not compiled with Unicode property
921            support
922      68  \ec must be followed by an ASCII character
923      69  \ek is not followed by a braced, angle-bracketed, or quoted name
924      70  internal error: unknown opcode in find_fixedlength()
925      71  \eN is not supported in a class
926      72  too many forward references
927      73  disallowed Unicode code point (>= 0xd800 && <= 0xdfff)
928      74  invalid UTF-16 string (specifically UTF-16)
929      75  name is too long in (*MARK), (*PRUNE), (*SKIP), or (*THEN)
930      76  character value in \eu.... sequence is too large
931  .sp  .sp
932  The numbers 32 and 10000 in errors 48 and 49 are defaults; different values may  The numbers 32 and 10000 in errors 48 and 49 are defaults; different values may
933  be used if the limits were changed when PCRE was built.  be used if the limits were changed when PCRE was built.
934  .  .
935  .  .
936    .\" HTML <a name="studyingapattern"></a>
937  .SH "STUDYING A PATTERN"  .SH "STUDYING A PATTERN"
938  .rs  .rs
939  .sp  .sp
# Line 818  If studying the pattern does not produce Line 964  If studying the pattern does not produce
964  wants to pass any of the other fields to \fBpcre_exec()\fP or  wants to pass any of the other fields to \fBpcre_exec()\fP or
965  \fBpcre_dfa_exec()\fP, it must set up its own \fBpcre_extra\fP block.  \fBpcre_dfa_exec()\fP, it must set up its own \fBpcre_extra\fP block.
966  .P  .P
967  The second argument of \fBpcre_study()\fP contains option bits. At present, no  The second argument of \fBpcre_study()\fP contains option bits. There are three
968  options are defined, and this argument should always be zero.  options:
969    .sp
970      PCRE_STUDY_JIT_COMPILE
971      PCRE_STUDY_JIT_PARTIAL_HARD_COMPILE
972      PCRE_STUDY_JIT_PARTIAL_SOFT_COMPILE
973    .sp
974    If any of these are set, and the just-in-time compiler is available, the
975    pattern is further compiled into machine code that executes much faster than
976    the \fBpcre_exec()\fP interpretive matching function. If the just-in-time
977    compiler is not available, these options are ignored. All other bits in the
978    \fIoptions\fP argument must be zero.
979    .P
980    JIT compilation is a heavyweight optimization. It can take some time for
981    patterns to be analyzed, and for one-off matches and simple patterns the
982    benefit of faster execution might be offset by a much slower study time.
983    Not all patterns can be optimized by the JIT compiler. For those that cannot be
984    handled, matching automatically falls back to the \fBpcre_exec()\fP
985    interpreter. For more details, see the
986    .\" HREF
987    \fBpcrejit\fP
988    .\"
989    documentation.
990  .P  .P
991  The third argument for \fBpcre_study()\fP is a pointer for an error message. If  The third argument for \fBpcre_study()\fP is a pointer for an error message. If
992  studying succeeds (even if no data is returned), the variable it points to is  studying succeeds (even if no data is returned), the variable it points to is
# Line 828  static string that is part of the librar Line 995  static string that is part of the librar
995  should test the error pointer for NULL after calling \fBpcre_study()\fP, to be  should test the error pointer for NULL after calling \fBpcre_study()\fP, to be
996  sure that it has run successfully.  sure that it has run successfully.
997  .P  .P
998  This is a typical call to \fBpcre_study\fP():  When you are finished with a pattern, you can free the memory used for the
999    study data by calling \fBpcre_free_study()\fP. This function was added to the
1000    API for release 8.20. For earlier versions, the memory could be freed with
1001    \fBpcre_free()\fP, just like the pattern itself. This will still work in cases
1002    where JIT optimization is not used, but it is advisable to change to the new
1003    function when convenient.
1004    .P
1005    This is a typical way in which \fBpcre_study\fP() is used (except that in a
1006    real application there should be tests for errors):
1007  .sp  .sp
1008    pcre_extra *pe;    int rc;
1009    pe = pcre_study(    pcre *re;
1010      pcre_extra *sd;
1011      re = pcre_compile("pattern", 0, &error, &erroroffset, NULL);
1012      sd = pcre_study(
1013      re,             /* result of pcre_compile() */      re,             /* result of pcre_compile() */
1014      0,              /* no options exist */      0,              /* no options */
1015      &error);        /* set to NULL or points to a message */      &error);        /* set to NULL or points to a message */
1016      rc = pcre_exec(   /* see below for details of pcre_exec() options */
1017        re, sd, "subject", 7, 0, 0, ovector, 30);
1018      ...
1019      pcre_free_study(sd);
1020      pcre_free(re);
1021  .sp  .sp
1022  Studying a pattern does two things: first, a lower bound for the length of  Studying a pattern does two things: first, a lower bound for the length of
1023  subject string that is needed to match the pattern is computed. This does not  subject string that is needed to match the pattern is computed. This does not
# Line 847  in a calling program via the \fBpcre_ful Line 1030  in a calling program via the \fBpcre_ful
1030  Studying a pattern is also useful for non-anchored patterns that do not have a  Studying a pattern is also useful for non-anchored patterns that do not have a
1031  single fixed starting character. A bitmap of possible starting bytes is  single fixed starting character. A bitmap of possible starting bytes is
1032  created. This speeds up finding a position in the subject at which to start  created. This speeds up finding a position in the subject at which to start
1033  matching.  matching. (In 16-bit mode, the bitmap is used for 16-bit values less than 256.)
1034  .P  .P
1035  The two optimizations just described can be disabled by setting the  These two optimizations apply to both \fBpcre_exec()\fP and
1036  PCRE_NO_START_OPTIMIZE option when calling \fBpcre_exec()\fP or  \fBpcre_dfa_exec()\fP, and the information is also used by the JIT compiler.
1037  \fBpcre_dfa_exec()\fP. You might want to do this if your pattern contains  The optimizations can be disabled by setting the PCRE_NO_START_OPTIMIZE option
1038  callouts, or make use of (*MARK), and you make use of these in cases where  when calling \fBpcre_exec()\fP or \fBpcre_dfa_exec()\fP, but if this is done,
1039  matching fails. See the discussion of PCRE_NO_START_OPTIMIZE  JIT execution is also disabled. You might want to do this if your pattern
1040    contains callouts or (*MARK) and you want to make use of these facilities in
1041    cases where matching fails. See the discussion of PCRE_NO_START_OPTIMIZE
1042  .\" HTML <a href="#execoptions">  .\" HTML <a href="#execoptions">
1043  .\" </a>  .\" </a>
1044  below.  below.
# Line 866  below. Line 1051  below.
1051  .sp  .sp
1052  PCRE handles caseless matching, and determines whether characters are letters,  PCRE handles caseless matching, and determines whether characters are letters,
1053  digits, or whatever, by reference to a set of tables, indexed by character  digits, or whatever, by reference to a set of tables, indexed by character
1054  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
1055  less than 128. By default, higher-valued codes never match escapes such as \ew  with codes less than 128. By default, higher-valued codes never match escapes
1056  or \ed, but they can be tested with \ep if PCRE is built with Unicode character  such as \ew or \ed, but they can be tested with \ep if PCRE is built with
1057  property support. Alternatively, the PCRE_UCP option can be set at compile  Unicode character property support. Alternatively, the PCRE_UCP option can be
1058  time; this causes \ew and friends to use Unicode property support instead of  set at compile time; this causes \ew and friends to use Unicode property
1059  built-in tables. The use of locales with Unicode is discouraged. If you are  support instead of built-in tables. The use of locales with Unicode is
1060  handling characters with codes greater than 128, you should either use UTF-8  discouraged. If you are handling characters with codes greater than 128, you
1061  and Unicode, or use locales, but not try to mix the two.  should either use UTF-8 and Unicode, or use locales, but not try to mix the
1062    two.
1063  .P  .P
1064  PCRE contains an internal set of tables that are used when the final argument  PCRE contains an internal set of tables that are used when the final argument
1065  of \fBpcre_compile()\fP is NULL. These are sufficient for many applications.  of \fBpcre_compile()\fP is NULL. These are sufficient for many applications.
# Line 918  one in which it was compiled. Passing ta Line 1104  one in which it was compiled. Passing ta
1104  below in the section on matching a pattern.  below in the section on matching a pattern.
1105  .  .
1106  .  .
1107    .\" HTML <a name="infoaboutpattern"></a>
1108  .SH "INFORMATION ABOUT A PATTERN"  .SH "INFORMATION ABOUT A PATTERN"
1109  .rs  .rs
1110  .sp  .sp
# Line 926  below in the section on matching a patte Line 1113  below in the section on matching a patte
1113  .B int \fIwhat\fP, void *\fIwhere\fP);  .B int \fIwhat\fP, void *\fIwhere\fP);
1114  .PP  .PP
1115  The \fBpcre_fullinfo()\fP function returns information about a compiled  The \fBpcre_fullinfo()\fP function returns information about a compiled
1116  pattern. It replaces the obsolete \fBpcre_info()\fP function, which is  pattern. It replaces the \fBpcre_info()\fP function, which was removed from the
1117  nevertheless retained for backwards compability (and is documented below).  library at version 8.30, after more than 10 years of obsolescence.
1118  .P  .P
1119  The first argument for \fBpcre_fullinfo()\fP is a pointer to the compiled  The first argument for \fBpcre_fullinfo()\fP is a pointer to the compiled
1120  pattern. The second argument is the result of \fBpcre_study()\fP, or NULL if  pattern. The second argument is the result of \fBpcre_study()\fP, or NULL if
# Line 936  information is required, and the fourth Line 1123  information is required, and the fourth
1123  to receive the data. The yield of the function is zero for success, or one of  to receive the data. The yield of the function is zero for success, or one of
1124  the following negative numbers:  the following negative numbers:
1125  .sp  .sp
1126    PCRE_ERROR_NULL       the argument \fIcode\fP was NULL    PCRE_ERROR_NULL           the argument \fIcode\fP was NULL
1127                          the argument \fIwhere\fP was NULL                              the argument \fIwhere\fP was NULL
1128    PCRE_ERROR_BADMAGIC   the "magic number" was not found    PCRE_ERROR_BADMAGIC       the "magic number" was not found
1129    PCRE_ERROR_BADOPTION  the value of \fIwhat\fP was invalid    PCRE_ERROR_BADENDIANNESS  the pattern was compiled with different
1130                                endianness
1131      PCRE_ERROR_BADOPTION      the value of \fIwhat\fP was invalid
1132  .sp  .sp
1133  The "magic number" is placed at the start of each compiled pattern as an simple  The "magic number" is placed at the start of each compiled pattern as an simple
1134  check against passing an arbitrary memory pointer. Here is a typical call of  check against passing an arbitrary memory pointer. The endianness error can
1135  \fBpcre_fullinfo()\fP, to obtain the length of the compiled pattern:  occur if a compiled pattern is saved and reloaded on a different host. Here is
1136    a typical call of \fBpcre_fullinfo()\fP, to obtain the length of the compiled
1137    pattern:
1138  .sp  .sp
1139    int rc;    int rc;
1140    size_t length;    size_t length;
1141    rc = pcre_fullinfo(    rc = pcre_fullinfo(
1142      re,               /* result of pcre_compile() */      re,               /* result of pcre_compile() */
1143      pe,               /* result of pcre_study(), or NULL */      sd,               /* result of pcre_study(), or NULL */
1144      PCRE_INFO_SIZE,   /* what is required */      PCRE_INFO_SIZE,   /* what is required */
1145      &length);         /* where to put the data */      &length);         /* where to put the data */
1146  .sp  .sp
# Line 977  a NULL table pointer. Line 1168  a NULL table pointer.
1168  .sp  .sp
1169    PCRE_INFO_FIRSTBYTE    PCRE_INFO_FIRSTBYTE
1170  .sp  .sp
1171  Return information about the first byte of any matched string, for a  Return information about the first data unit of any matched string, for a
1172  non-anchored pattern. The fourth argument should point to an \fBint\fP  non-anchored pattern. (The name of this option refers to the 8-bit library,
1173  variable. (This option used to be called PCRE_INFO_FIRSTCHAR; the old name is  where data units are bytes.) The fourth argument should point to an \fBint\fP
1174  still recognized for backwards compatibility.)  variable.
1175    .P
1176    If there is a fixed first value, for example, the letter "c" from a pattern
1177    such as (cat|cow|coyote), its value is returned. In the 8-bit library, the
1178    value is always less than 256; in the 16-bit library the value can be up to
1179    0xffff.
1180  .P  .P
1181  If there is a fixed first byte, for example, from a pattern such as  If there is no fixed first value, and if either
 (cat|cow|coyote), its value is returned. Otherwise, if either  
1182  .sp  .sp
1183  (a) the pattern was compiled with the PCRE_MULTILINE option, and every branch  (a) the pattern was compiled with the PCRE_MULTILINE option, and every branch
1184  starts with "^", or  starts with "^", or
# Line 998  returned. For anchored patterns, -2 is r Line 1193  returned. For anchored patterns, -2 is r
1193    PCRE_INFO_FIRSTTABLE    PCRE_INFO_FIRSTTABLE
1194  .sp  .sp
1195  If the pattern was studied, and this resulted in the construction of a 256-bit  If the pattern was studied, and this resulted in the construction of a 256-bit
1196  table indicating a fixed set of bytes for the first byte in any matching  table indicating a fixed set of values for the first data unit in any matching
1197  string, a pointer to the table is returned. Otherwise NULL is returned. The  string, a pointer to the table is returned. Otherwise NULL is returned. The
1198  fourth argument should point to an \fBunsigned char *\fP variable.  fourth argument should point to an \fBunsigned char *\fP variable.
1199  .sp  .sp
# Line 1014  Return 1 if the (?J) or (?-J) option set Line 1209  Return 1 if the (?J) or (?-J) option set
1209  0. The fourth argument should point to an \fBint\fP variable. (?J) and  0. The fourth argument should point to an \fBint\fP variable. (?J) and
1210  (?-J) set and unset the local PCRE_DUPNAMES option, respectively.  (?-J) set and unset the local PCRE_DUPNAMES option, respectively.
1211  .sp  .sp
1212      PCRE_INFO_JIT
1213    .sp
1214    Return 1 if the pattern was studied with one of the JIT options, and
1215    just-in-time compiling was successful. The fourth argument should point to an
1216    \fBint\fP variable. A return value of 0 means that JIT support is not available
1217    in this version of PCRE, or that the pattern was not studied with a JIT option,
1218    or that the JIT compiler could not handle this particular pattern. See the
1219    .\" HREF
1220    \fBpcrejit\fP
1221    .\"
1222    documentation for details of what can and cannot be handled.
1223    .sp
1224      PCRE_INFO_JITSIZE
1225    .sp
1226    If the pattern was successfully studied with a JIT option, return the size of
1227    the JIT compiled code, otherwise return zero. The fourth argument should point
1228    to a \fBsize_t\fP variable.
1229    .sp
1230    PCRE_INFO_LASTLITERAL    PCRE_INFO_LASTLITERAL
1231  .sp  .sp
1232  Return the value of the rightmost literal byte that must exist in any matched  Return the value of the rightmost literal data unit that must exist in any
1233  string, other than at its start, if such a byte has been recorded. The fourth  matched string, other than at its start, if such a value has been recorded. The
1234  argument should point to an \fBint\fP variable. If there is no such byte, -1 is  fourth argument should point to an \fBint\fP variable. If there is no such
1235  returned. For anchored patterns, a last literal byte is recorded only if it  value, -1 is returned. For anchored patterns, a last literal value is recorded
1236  follows something of variable length. For example, for the pattern  only if it follows something of variable length. For example, for the pattern
1237  /^a\ed+z\ed+/ the returned value is "z", but for /^a\edz\ed/ the returned value  /^a\ed+z\ed+/ the returned value is "z", but for /^a\edz\ed/ the returned value
1238  is -1.  is -1.
1239  .sp  .sp
1240      PCRE_INFO_MAXLOOKBEHIND
1241    .sp
1242    Return the number of characters (NB not bytes) in the longest lookbehind
1243    assertion in the pattern. Note that the simple assertions \eb and \eB require a
1244    one-character lookbehind. This information is useful when doing multi-segment
1245    matching using the partial matching facilities.
1246    .sp
1247    PCRE_INFO_MINLENGTH    PCRE_INFO_MINLENGTH
1248  .sp  .sp
1249  If the pattern was studied and a minimum length for matching subject strings  If the pattern was studied and a minimum length for matching subject strings
1250  was computed, its value is returned. Otherwise the returned value is -1. The  was computed, its value is returned. Otherwise the returned value is -1. The
1251  value is a number of characters, not bytes (this may be relevant in UTF-8  value is a number of characters, which in UTF-8 mode may be different from the
1252  mode). The fourth argument should point to an \fBint\fP variable. A  number of bytes. The fourth argument should point to an \fBint\fP variable. A
1253  non-negative value is a lower bound to the length of any matching string. There  non-negative value is a lower bound to the length of any matching string. There
1254  may not be any strings of that length that do actually match, but every string  may not be any strings of that length that do actually match, but every string
1255  that does match is at least that long.  that does match is at least that long.
# Line 1052  The map consists of a number of fixed-si Line 1272  The map consists of a number of fixed-si
1272  the number of entries, and PCRE_INFO_NAMEENTRYSIZE gives the size of each  the number of entries, and PCRE_INFO_NAMEENTRYSIZE gives the size of each
1273  entry; both of these return an \fBint\fP value. The entry size depends on the  entry; both of these return an \fBint\fP value. The entry size depends on the
1274  length of the longest name. PCRE_INFO_NAMETABLE returns a pointer to the first  length of the longest name. PCRE_INFO_NAMETABLE returns a pointer to the first
1275  entry of the table (a pointer to \fBchar\fP). The first two bytes of each entry  entry of the table. This is a pointer to \fBchar\fP in the 8-bit library, where
1276  are the number of the capturing parenthesis, most significant byte first. The  the first two bytes of each entry are the number of the capturing parenthesis,
1277  rest of the entry is the corresponding name, zero terminated.  most significant byte first. In the 16-bit library, the pointer points to
1278    16-bit data units, the first of which contains the parenthesis number. The rest
1279    of the entry is the corresponding name, zero terminated.
1280  .P  .P
1281  The names are in alphabetical order. Duplicate names may appear if (?| is used  The names are in alphabetical order. Duplicate names may appear if (?| is used
1282  to create multiple groups with the same number, as described in the  to create multiple groups with the same number, as described in the
# Line 1073  table in the order in which they were fo Line 1295  table in the order in which they were fo
1295  necessarily the case because later subpatterns may have lower numbers.  necessarily the case because later subpatterns may have lower numbers.
1296  .P  .P
1297  As a simple example of the name/number table, consider the following pattern  As a simple example of the name/number table, consider the following pattern
1298  (assume PCRE_EXTENDED is set, so white space - including newlines - is  after compilation by the 8-bit library (assume PCRE_EXTENDED is set, so white
1299  ignored):  space - including newlines - is ignored):
1300  .sp  .sp
1301  .\" JOIN  .\" JOIN
1302    (?<date> (?<year>(\ed\ed)?\ed\ed) -    (?<date> (?<year>(\ed\ed)?\ed\ed) -
# Line 1129  For such patterns, the PCRE_ANCHORED bit Line 1351  For such patterns, the PCRE_ANCHORED bit
1351  .sp  .sp
1352    PCRE_INFO_SIZE    PCRE_INFO_SIZE
1353  .sp  .sp
1354  Return the size of the compiled pattern, that is, the value that was passed as  Return the size of the compiled pattern in bytes (for both libraries). The
1355  the argument to \fBpcre_malloc()\fP when PCRE was getting memory in which to  fourth argument should point to a \fBsize_t\fP variable. This value does not
1356  place the compiled data. The fourth argument should point to a \fBsize_t\fP  include the size of the \fBpcre\fP structure that is returned by
1357  variable.  \fBpcre_compile()\fP. The value that is passed as the argument to
1358    \fBpcre_malloc()\fP when \fBpcre_compile()\fP is getting memory in which to
1359    place the compiled data is the value returned by this option plus the size of
1360    the \fBpcre\fP structure. Studying a compiled pattern, with or without JIT,
1361    does not alter the value returned by this option.
1362  .sp  .sp
1363    PCRE_INFO_STUDYSIZE    PCRE_INFO_STUDYSIZE
1364  .sp  .sp
1365  Return the size of the data block pointed to by the \fIstudy_data\fP field in  Return the size in bytes of the data block pointed to by the \fIstudy_data\fP
1366  a \fBpcre_extra\fP block. That is, it is the value that was passed to  field in a \fBpcre_extra\fP block. If \fBpcre_extra\fP is NULL, or there is no
 \fBpcre_malloc()\fP when PCRE was getting memory into which to place the data  
 created by \fBpcre_study()\fP. If \fBpcre_extra\fP is NULL, or there is no  
1367  study data, zero is returned. The fourth argument should point to a  study data, zero is returned. The fourth argument should point to a
1368  \fBsize_t\fP variable.  \fBsize_t\fP variable. The \fIstudy_data\fP field is set by \fBpcre_study()\fP
1369  .  to record information that will speed up matching (see the section entitled
1370  .  .\" HTML <a href="#studyingapattern">
1371  .SH "OBSOLETE INFO FUNCTION"  .\" </a>
1372  .rs  "Studying a pattern"
1373  .sp  .\"
1374  .B int pcre_info(const pcre *\fIcode\fP, int *\fIoptptr\fP, int  above). The format of the \fIstudy_data\fP block is private, but its length
1375  .B *\fIfirstcharptr\fP);  is made available via this option so that it can be saved and restored (see the
1376  .PP  .\" HREF
1377  The \fBpcre_info()\fP function is now obsolete because its interface is too  \fBpcreprecompile\fP
1378  restrictive to return all the available data about a compiled pattern. New  .\"
1379  programs should use \fBpcre_fullinfo()\fP instead. The yield of  documentation for details).
 \fBpcre_info()\fP is the number of capturing subpatterns, or one of the  
 following negative numbers:  
 .sp  
   PCRE_ERROR_NULL       the argument \fIcode\fP was NULL  
   PCRE_ERROR_BADMAGIC   the "magic number" was not found  
 .sp  
 If the \fIoptptr\fP argument is not NULL, a copy of the options with which the  
 pattern was compiled is placed in the integer it points to (see  
 PCRE_INFO_OPTIONS above).  
 .P  
 If the pattern is not anchored and the \fIfirstcharptr\fP argument is not NULL,  
 it is used to pass back information about the first character of any matched  
 string (see PCRE_INFO_FIRSTBYTE above).  
1380  .  .
1381  .  .
1382  .SH "REFERENCE COUNTS"  .SH "REFERENCE COUNTS"
# Line 1203  is different. (This seems a highly unlik Line 1414  is different. (This seems a highly unlik
1414  The function \fBpcre_exec()\fP is called to match a subject string against a  The function \fBpcre_exec()\fP is called to match a subject string against a
1415  compiled pattern, which is passed in the \fIcode\fP argument. If the  compiled pattern, which is passed in the \fIcode\fP argument. If the
1416  pattern was studied, the result of the study should be passed in the  pattern was studied, the result of the study should be passed in the
1417  \fIextra\fP argument. This function is the main matching facility of the  \fIextra\fP argument. You can call \fBpcre_exec()\fP with the same \fIcode\fP
1418  library, and it operates in a Perl-like manner. For specialist use there is  and \fIextra\fP arguments as many times as you like, in order to match
1419  also an alternative matching function, which is described  different subject strings with the same pattern.
1420    .P
1421    This function is the main matching facility of the library, and it operates in
1422    a Perl-like manner. For specialist use there is also an alternative matching
1423    function, which is described
1424  .\" HTML <a href="#dfamatch">  .\" HTML <a href="#dfamatch">
1425  .\" </a>  .\" </a>
1426  below  below
# Line 1236  Here is an example of a simple call to \ Line 1451  Here is an example of a simple call to \
1451      ovector,        /* vector of integers for substring information */      ovector,        /* vector of integers for substring information */
1452      30);            /* number of elements (NOT size in bytes) */      30);            /* number of elements (NOT size in bytes) */
1453  .  .
1454    .
1455  .\" HTML <a name="extradata"></a>  .\" HTML <a name="extradata"></a>
1456  .SS "Extra data for \fBpcre_exec()\fR"  .SS "Extra data for \fBpcre_exec()\fR"
1457  .rs  .rs
# Line 1248  fields (not necessarily in this order): Line 1464  fields (not necessarily in this order):
1464  .sp  .sp
1465    unsigned long int \fIflags\fP;    unsigned long int \fIflags\fP;
1466    void *\fIstudy_data\fP;    void *\fIstudy_data\fP;
1467      void *\fIexecutable_jit\fP;
1468    unsigned long int \fImatch_limit\fP;    unsigned long int \fImatch_limit\fP;
1469    unsigned long int \fImatch_limit_recursion\fP;    unsigned long int \fImatch_limit_recursion\fP;
1470    void *\fIcallout_data\fP;    void *\fIcallout_data\fP;
1471    const unsigned char *\fItables\fP;    const unsigned char *\fItables\fP;
1472    unsigned char **\fImark\fP;    unsigned char **\fImark\fP;
1473  .sp  .sp
1474  The \fIflags\fP field is a bitmap that specifies which of the other fields  In the 16-bit version of this structure, the \fImark\fP field has type
1475  are set. The flag bits are:  "PCRE_UCHAR16 **".
1476    .P
1477    The \fIflags\fP field is used to specify which of the other fields are set. The
1478    flag bits are:
1479  .sp  .sp
1480    PCRE_EXTRA_STUDY_DATA    PCRE_EXTRA_CALLOUT_DATA
1481      PCRE_EXTRA_EXECUTABLE_JIT
1482      PCRE_EXTRA_MARK
1483    PCRE_EXTRA_MATCH_LIMIT    PCRE_EXTRA_MATCH_LIMIT
1484    PCRE_EXTRA_MATCH_LIMIT_RECURSION    PCRE_EXTRA_MATCH_LIMIT_RECURSION
1485    PCRE_EXTRA_CALLOUT_DATA    PCRE_EXTRA_STUDY_DATA
1486    PCRE_EXTRA_TABLES    PCRE_EXTRA_TABLES
   PCRE_EXTRA_MARK  
1487  .sp  .sp
1488  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
1489  \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
1490  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
1491  the block by setting the other fields and their corresponding flag bits.  should not set these yourself, but you may add to the block by setting other
1492    fields and their corresponding flag bits.
1493  .P  .P
1494  The \fImatch_limit\fP field provides a means of preventing PCRE from using up a  The \fImatch_limit\fP field provides a means of preventing PCRE from using up a
1495  vast amount of resources when running patterns that are not going to match,  vast amount of resources when running patterns that are not going to match,
1496  but which have a very large number of possibilities in their search trees. The  but which have a very large number of possibilities in their search trees. The
1497  classic example is a pattern that uses nested unlimited repeats.  classic example is a pattern that uses nested unlimited repeats.
1498  .P  .P
1499  Internally, PCRE uses a function called \fBmatch()\fP which it calls repeatedly  Internally, \fBpcre_exec()\fP uses a function called \fBmatch()\fP, which it
1500  (sometimes recursively). The limit set by \fImatch_limit\fP is imposed on the  calls repeatedly (sometimes recursively). The limit set by \fImatch_limit\fP is
1501  number of times this function is called during a match, which has the effect of  imposed on the number of times this function is called during a match, which
1502  limiting the amount of backtracking that can take place. For patterns that are  has the effect of limiting the amount of backtracking that can take place. For
1503  not anchored, the count restarts from zero for each position in the subject  patterns that are not anchored, the count restarts from zero for each position
1504  string.  in the subject string.
1505    .P
1506    When \fBpcre_exec()\fP is called with a pattern that was successfully studied
1507    with a JIT option, the way that the matching is executed is entirely different.
1508    However, there is still the possibility of runaway matching that goes on for a
1509    very long time, and so the \fImatch_limit\fP value is also used in this case
1510    (but in a different way) to limit how long the matching can continue.
1511  .P  .P
1512  The default value for the limit can be set when PCRE is built; the default  The default value for the limit can be set when PCRE is built; the default
1513  default is 10 million, which handles all but the most extreme cases. You can  default is 10 million, which handles all but the most extreme cases. You can
# Line 1294  limits the depth of recursion. The recur Line 1522  limits the depth of recursion. The recur
1522  total number of calls, because not all calls to \fBmatch()\fP are recursive.  total number of calls, because not all calls to \fBmatch()\fP are recursive.
1523  This limit is of use only if it is set smaller than \fImatch_limit\fP.  This limit is of use only if it is set smaller than \fImatch_limit\fP.
1524  .P  .P
1525  Limiting the recursion depth limits the amount of stack that can be used, or,  Limiting the recursion depth limits the amount of machine stack that can be
1526  when PCRE has been compiled to use memory on the heap instead of the stack, the  used, or, when PCRE has been compiled to use memory on the heap instead of the
1527  amount of heap memory that can be used.  stack, the amount of heap memory that can be used. This limit is not relevant,
1528    and is ignored, when matching is done using JIT compiled code.
1529  .P  .P
1530  The default value for \fImatch_limit_recursion\fP can be set when PCRE is  The default value for \fImatch_limit_recursion\fP can be set when PCRE is
1531  built; the default default is the same value as the default for  built; the default default is the same value as the default for
# Line 1327  called. See the Line 1556  called. See the
1556  documentation for a discussion of saving compiled patterns for later use.  documentation for a discussion of saving compiled patterns for later use.
1557  .P  .P
1558  If PCRE_EXTRA_MARK is set in the \fIflags\fP field, the \fImark\fP field must  If PCRE_EXTRA_MARK is set in the \fIflags\fP field, the \fImark\fP field must
1559  be set to point to a \fBchar *\fP variable. If the pattern contains any  be set to point to a suitable variable. If the pattern contains any
1560  backtracking control verbs such as (*MARK:NAME), and the execution ends up with  backtracking control verbs such as (*MARK:NAME), and the execution ends up with
1561  a name to pass back, a pointer to the name string (zero terminated) is placed  a name to pass back, a pointer to the name string (zero terminated) is placed
1562  in the variable pointed to by the \fImark\fP field. The names are within the  in the variable pointed to by the \fImark\fP field. The names are within the
1563  compiled pattern; if you wish to retain such a name you must copy it before  compiled pattern; if you wish to retain such a name you must copy it before
1564  freeing the memory of a compiled pattern. If there is no name to pass back, the  freeing the memory of a compiled pattern. If there is no name to pass back, the
1565  variable pointed to by the \fImark\fP field set to NULL. For details of the  variable pointed to by the \fImark\fP field is set to NULL. For details of the
1566  backtracking control verbs, see the section entitled  backtracking control verbs, see the section entitled
1567  .\" HTML <a href="pcrepattern#backtrackcontrol">  .\" HTML <a href="pcrepattern#backtrackcontrol">
1568  .\" </a>  .\" </a>
# Line 1353  documentation. Line 1582  documentation.
1582  The unused bits of the \fIoptions\fP argument for \fBpcre_exec()\fP must be  The unused bits of the \fIoptions\fP argument for \fBpcre_exec()\fP must be
1583  zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_\fIxxx\fP,  zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_\fIxxx\fP,
1584  PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NOTEMPTY_ATSTART,  PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NOTEMPTY_ATSTART,
1585  PCRE_NO_START_OPTIMIZE, PCRE_NO_UTF8_CHECK, PCRE_PARTIAL_SOFT, and  PCRE_NO_START_OPTIMIZE, PCRE_NO_UTF8_CHECK, PCRE_PARTIAL_HARD, and
1586  PCRE_PARTIAL_HARD.  PCRE_PARTIAL_SOFT.
1587    .P
1588    If the pattern was successfully studied with one of the just-in-time (JIT)
1589    compile options, the only supported options for JIT execution are
1590    PCRE_NO_UTF8_CHECK, PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY,
1591    PCRE_NOTEMPTY_ATSTART, PCRE_PARTIAL_HARD, and PCRE_PARTIAL_SOFT. If an
1592    unsupported option is used, JIT execution is disabled and the normal
1593    interpretive code in \fBpcre_exec()\fP is run.
1594  .sp  .sp
1595    PCRE_ANCHORED    PCRE_ANCHORED
1596  .sp  .sp
# Line 1450  the Line 1686  the
1686  .\" HREF  .\" HREF
1687  \fBpcredemo\fP  \fBpcredemo\fP
1688  .\"  .\"
1689  sample program. In the most general case, you have to check to see if the  sample program. In the most general case, you have to check to see if the
1690  newline convention recognizes CRLF as a newline, and if so, and the current  newline convention recognizes CRLF as a newline, and if so, and the current
1691  character is CR followed by LF, advance the starting offset by two characters  character is CR followed by LF, advance the starting offset by two characters
1692  instead of one.  instead of one.
1693  .sp  .sp
# Line 1471  a pre-scan of the subject that takes pla Line 1707  a pre-scan of the subject that takes pla
1707  The PCRE_NO_START_OPTIMIZE option disables the start-up optimizations, possibly  The PCRE_NO_START_OPTIMIZE option disables the start-up optimizations, possibly
1708  causing performance to suffer, but ensuring that in cases where the result is  causing performance to suffer, but ensuring that in cases where the result is
1709  "no match", the callouts do occur, and that items such as (*COMMIT) and (*MARK)  "no match", the callouts do occur, and that items such as (*COMMIT) and (*MARK)
1710  are considered at every possible starting position in the subject string.  are considered at every possible starting position in the subject string. If
1711    PCRE_NO_START_OPTIMIZE is set at compile time, it cannot be unset at matching
1712    time. The use of PCRE_NO_START_OPTIMIZE disables JIT execution; when it is set,
1713    matching is always done using interpretively.
1714    .P
1715  Setting PCRE_NO_START_OPTIMIZE can change the outcome of a matching operation.  Setting PCRE_NO_START_OPTIMIZE can change the outcome of a matching operation.
1716  Consider the pattern  Consider the pattern
1717  .sp  .sp
# Line 1503  returned. Line 1743  returned.
1743  .sp  .sp
1744  When PCRE_UTF8 is set at compile time, the validity of the subject as a UTF-8  When PCRE_UTF8 is set at compile time, the validity of the subject as a UTF-8
1745  string is automatically checked when \fBpcre_exec()\fP is subsequently called.  string is automatically checked when \fBpcre_exec()\fP is subsequently called.
1746  The value of \fIstartoffset\fP is also checked to ensure that it points to the  The entire string is checked before any other processing takes place. The value
1747  start of a UTF-8 character. There is a discussion about the validity of UTF-8  of \fIstartoffset\fP is also checked to ensure that it points to the start of a
1748  strings in the  UTF-8 character. There is a discussion about the
1749  .\" HTML <a href="pcre.html#utf8strings">  .\" HTML <a href="pcreunicode.html#utf8strings">
1750  .\" </a>  .\" </a>
1751  section on UTF-8 support  validity of UTF-8 strings
1752  .\"  .\"
1753  in the main  in the
1754  .\" HREF  .\" HREF
1755  \fBpcre\fP  \fBpcreunicode\fP
1756  .\"  .\"
1757  page. If an invalid UTF-8 sequence of bytes is found, \fBpcre_exec()\fP returns  page. If an invalid sequence of bytes is found, \fBpcre_exec()\fP returns the
1758  the error PCRE_ERROR_BADUTF8. If \fIstartoffset\fP contains a value that does  error PCRE_ERROR_BADUTF8 or, if PCRE_PARTIAL_HARD is set and the problem is a
1759  not point to the start of a UTF-8 character (or to the end of the subject),  truncated character at the end of the subject, PCRE_ERROR_SHORTUTF8. In both
1760  PCRE_ERROR_BADUTF8_OFFSET is returned.  cases, information about the precise nature of the error may also be returned
1761    (see the descriptions of these errors in the section entitled \fIError return
1762    values from\fP \fBpcre_exec()\fP
1763    .\" HTML <a href="#errorlist">
1764    .\" </a>
1765    below).
1766    .\"
1767    If \fIstartoffset\fP contains a value that does not point to the start of a
1768    UTF-8 character (or to the end of the subject), PCRE_ERROR_BADUTF8_OFFSET is
1769    returned.
1770  .P  .P
1771  If you already know that your subject is valid, and you want to skip these  If you already know that your subject is valid, and you want to skip these
1772  checks for performance reasons, you can set the PCRE_NO_UTF8_CHECK option when  checks for performance reasons, you can set the PCRE_NO_UTF8_CHECK option when
1773  calling \fBpcre_exec()\fP. You might want to do this for the second and  calling \fBpcre_exec()\fP. You might want to do this for the second and
1774  subsequent calls to \fBpcre_exec()\fP if you are making repeated calls to find  subsequent calls to \fBpcre_exec()\fP if you are making repeated calls to find
1775  all the matches in a single subject string. However, you should be sure that  all the matches in a single subject string. However, you should be sure that
1776  the value of \fIstartoffset\fP points to the start of a UTF-8 character (or the  the value of \fIstartoffset\fP points to the start of a character (or the end
1777  end of the subject). When PCRE_NO_UTF8_CHECK is set, the effect of passing an  of the subject). When PCRE_NO_UTF8_CHECK is set, the effect of passing an
1778  invalid UTF-8 string as a subject or an invalid value of \fIstartoffset\fP is  invalid string as a subject or an invalid value of \fIstartoffset\fP is
1779  undefined. Your program may crash.  undefined. Your program may crash.
1780  .sp  .sp
1781    PCRE_PARTIAL_HARD    PCRE_PARTIAL_HARD
# Line 1545  but only if no complete match can be fou Line 1794  but only if no complete match can be fou
1794  If PCRE_PARTIAL_HARD is set, it overrides PCRE_PARTIAL_SOFT. In this case, if a  If PCRE_PARTIAL_HARD is set, it overrides PCRE_PARTIAL_SOFT. In this case, if a
1795  partial match is found, \fBpcre_exec()\fP immediately returns  partial match is found, \fBpcre_exec()\fP immediately returns
1796  PCRE_ERROR_PARTIAL, without considering any other alternatives. In other words,  PCRE_ERROR_PARTIAL, without considering any other alternatives. In other words,
1797  when PCRE_PARTIAL_HARD is set, a partial match is considered to be more  when PCRE_PARTIAL_HARD is set, a partial match is considered to be more
1798  important that an alternative complete match.  important that an alternative complete match.
1799  .P  .P
1800  In both cases, the portion of the string that was inspected when the partial  In both cases, the portion of the string that was inspected when the partial
# Line 1561  documentation. Line 1810  documentation.
1810  .rs  .rs
1811  .sp  .sp
1812  The subject string is passed to \fBpcre_exec()\fP as a pointer in  The subject string is passed to \fBpcre_exec()\fP as a pointer in
1813  \fIsubject\fP, a length (in bytes) in \fIlength\fP, and a starting byte offset  \fIsubject\fP, a length in bytes in \fIlength\fP, and a starting byte offset
1814  in \fIstartoffset\fP. If this is negative or greater than the length of the  in \fIstartoffset\fP. If this is negative or greater than the length of the
1815  subject, \fBpcre_exec()\fP returns PCRE_ERROR_BADOFFSET.  subject, \fBpcre_exec()\fP returns PCRE_ERROR_BADOFFSET. When the starting
1816  .P  offset is zero, the search for a match starts at the beginning of the subject,
1817  In UTF-8 mode, the byte offset must point to the start of a UTF-8 character (or  and this is by far the most common case. In UTF-8 mode, the byte offset must
1818  the end of the subject). Unlike the pattern string, the subject may contain  point to the start of a UTF-8 character (or the end of the subject). Unlike the
1819  binary zero bytes. When the starting offset is zero, the search for a match  pattern string, the subject may contain binary zero bytes.
 starts at the beginning of the subject, and this is by far the most common  
 case.  
1820  .P  .P
1821  A non-zero starting offset is useful when searching for another match in the  A non-zero starting offset is useful when searching for another match in the
1822  same subject by calling \fBpcre_exec()\fP again after a previous success.  same subject by calling \fBpcre_exec()\fP again after a previous success.
# Line 1598  do this in the Line 1845  do this in the
1845  .\" HREF  .\" HREF
1846  \fBpcredemo\fP  \fBpcredemo\fP
1847  .\"  .\"
1848  sample program. In the most general case, you have to check to see if the  sample program. In the most general case, you have to check to see if the
1849  newline convention recognizes CRLF as a newline, and if so, and the current  newline convention recognizes CRLF as a newline, and if so, and the current
1850  character is CR followed by LF, advance the starting offset by two characters  character is CR followed by LF, advance the starting offset by two characters
1851  instead of one.  instead of one.
1852  .P  .P
# Line 1651  string that it matched that is returned. Line 1898  string that it matched that is returned.
1898  .P  .P
1899  If the vector is too small to hold all the captured substring offsets, it is  If the vector is too small to hold all the captured substring offsets, it is
1900  used as far as possible (up to two-thirds of its length), and the function  used as far as possible (up to two-thirds of its length), and the function
1901  returns a value of zero. If the substring offsets are not of interest,  returns a value of zero. If neither the actual string matched nor any captured
1902  \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
1903  \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
1904  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
1905  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
1906  advisable to supply an \fIovector\fP.  is usually advisable to supply an \fIovector\fP of reasonable size.
1907    .P
1908    There are some cases where zero is returned (indicating vector overflow) when
1909    in fact the vector is exactly the right size for the final match. For example,
1910    consider the pattern
1911    .sp
1912      (a)(?:(b)c|bd)
1913    .sp
1914    If a vector of 6 elements (allowing for only 1 captured substring) is given
1915    with subject string "abd", \fBpcre_exec()\fP will try to set the second
1916    captured string, thereby recording a vector overflow, before failing to match
1917    "c" and backing up to try the second alternative. The zero return, however,
1918    does correctly indicate that the maximum number of slots (namely 2) have been
1919    filled. In similar cases where there is temporary overflow, but the final
1920    number of used slots is actually less than the maximum, a non-zero value is
1921    returned.
1922  .P  .P
1923  The \fBpcre_fullinfo()\fP function can be used to find out how many capturing  The \fBpcre_fullinfo()\fP function can be used to find out how many capturing
1924  subpatterns there are in a compiled pattern. The smallest size for  subpatterns there are in a compiled pattern. The smallest size for
# Line 1677  return from the function is 2, because t Line 1939  return from the function is 2, because t
1939  number is 1, and the offsets for for the second and third capturing subpatterns  number is 1, and the offsets for for the second and third capturing subpatterns
1940  (assuming the vector is large enough, of course) are set to -1.  (assuming the vector is large enough, of course) are set to -1.
1941  .P  .P
1942  \fBNote\fP: Elements of \fIovector\fP that do not correspond to capturing  \fBNote\fP: Elements in the first two-thirds of \fIovector\fP that do not
1943  parentheses in the pattern are never changed. That is, if a pattern contains  correspond to capturing parentheses in the pattern are never changed. That is,
1944  \fIn\fP capturing parentheses, no more than \fIovector[0]\fP to  if a pattern contains \fIn\fP capturing parentheses, no more than
1945  \fIovector[2n+1]\fP are set by \fBpcre_exec()\fP. The other elements retain  \fIovector[0]\fP to \fIovector[2n+1]\fP are set by \fBpcre_exec()\fP. The other
1946  whatever values they previously had.  elements (in the first two-thirds) retain whatever values they previously had.
1947  .P  .P
1948  Some convenience functions are provided for extracting the captured substrings  Some convenience functions are provided for extracting the captured substrings
1949  as separate strings. These are described below.  as separate strings. These are described below.
1950  .  .
1951    .
1952  .\" HTML <a name="errorlist"></a>  .\" HTML <a name="errorlist"></a>
1953  .SS "Error return values from \fBpcre_exec()\fP"  .SS "Error return values from \fBpcre_exec()\fP"
1954  .rs  .rs
# Line 1755  documentation for details. Line 2018  documentation for details.
2018  .sp  .sp
2019    PCRE_ERROR_BADUTF8        (-10)    PCRE_ERROR_BADUTF8        (-10)
2020  .sp  .sp
2021  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,
2022    and the PCRE_NO_UTF8_CHECK option was not set. If the size of the output vector
2023    (\fIovecsize\fP) is at least 2, the byte offset to the start of the the invalid
2024    UTF-8 character is placed in the first element, and a reason code is placed in
2025    the second element. The reason codes are listed in the
2026    .\" HTML <a href="#badutf8reasons">
2027    .\" </a>
2028    following section.
2029    .\"
2030    For backward compatibility, if PCRE_PARTIAL_HARD is set and the problem is a
2031    truncated UTF-8 character at the end of the subject (reason codes 1 to 5),
2032    PCRE_ERROR_SHORTUTF8 is returned instead of PCRE_ERROR_BADUTF8.
2033  .sp  .sp
2034    PCRE_ERROR_BADUTF8_OFFSET (-11)    PCRE_ERROR_BADUTF8_OFFSET (-11)
2035  .sp  .sp
2036  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
2037  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
2038    \fIstartoffset\fP did not point to the beginning of a UTF-8 character or the
2039    end of the subject.
2040  .sp  .sp
2041    PCRE_ERROR_PARTIAL        (-12)    PCRE_ERROR_PARTIAL        (-12)
2042  .sp  .sp
# Line 1798  An invalid combination of PCRE_NEWLINE_\ Line 2074  An invalid combination of PCRE_NEWLINE_\
2074  .sp  .sp
2075    PCRE_ERROR_BADOFFSET      (-24)    PCRE_ERROR_BADOFFSET      (-24)
2076  .sp  .sp
2077  The value of \fIstartoffset\fP was negative or greater than the length of the  The value of \fIstartoffset\fP was negative or greater than the length of the
2078  subject, that is, the value in \fIlength\fP.  subject, that is, the value in \fIlength\fP.
2079    .sp
2080      PCRE_ERROR_SHORTUTF8      (-25)
2081    .sp
2082    This error is returned instead of PCRE_ERROR_BADUTF8 when the subject string
2083    ends with a truncated UTF-8 character and the PCRE_PARTIAL_HARD option is set.
2084    Information about the failure is returned as for PCRE_ERROR_BADUTF8. It is in
2085    fact sufficient to detect this case, but this special error code for
2086    PCRE_PARTIAL_HARD precedes the implementation of returned information; it is
2087    retained for backwards compatibility.
2088    .sp
2089      PCRE_ERROR_RECURSELOOP    (-26)
2090    .sp
2091    This error is returned when \fBpcre_exec()\fP detects a recursion loop within
2092    the pattern. Specifically, it means that either the whole pattern or a
2093    subpattern has been called recursively for the second time at the same position
2094    in the subject string. Some simple patterns that might do this are detected and
2095    faulted at compile time, but more complicated cases, in particular mutual
2096    recursions between two different subpatterns, cannot be detected until run
2097    time.
2098    .sp
2099      PCRE_ERROR_JIT_STACKLIMIT (-27)
2100    .sp
2101    This error is returned when a pattern that was successfully studied using a
2102    JIT compile option is being matched, but the memory available for the
2103    just-in-time processing stack is not large enough. See the
2104    .\" HREF
2105    \fBpcrejit\fP
2106    .\"
2107    documentation for more details.
2108    .sp
2109      PCRE_ERROR_BADMODE        (-28)
2110    .sp
2111    This error is given if a pattern that was compiled by the 8-bit library is
2112    passed to a 16-bit library function, or vice versa.
2113    .sp
2114      PCRE_ERROR_BADENDIANNESS  (-29)
2115    .sp
2116    This error is given if a pattern that was compiled and saved is reloaded on a
2117    host with different endianness. The utility function
2118    \fBpcre_pattern_to_host_byte_order()\fP can be used to convert such a pattern
2119    so that it runs on the new host.
2120  .P  .P
2121  Error numbers -16 to -20 and -22 are not used by \fBpcre_exec()\fP.  Error numbers -16 to -20, -22, and -30 are not used by \fBpcre_exec()\fP.
2122    .
2123    .
2124    .\" HTML <a name="badutf8reasons"></a>
2125    .SS "Reason codes for invalid UTF-8 strings"
2126    .rs
2127    .sp
2128    This section applies only to the 8-bit library. The corresponding information
2129    for the 16-bit library is given in the
2130    .\" HREF
2131    \fBpcre16\fP
2132    .\"
2133    page.
2134    .P
2135    When \fBpcre_exec()\fP returns either PCRE_ERROR_BADUTF8 or
2136    PCRE_ERROR_SHORTUTF8, and the size of the output vector (\fIovecsize\fP) is at
2137    least 2, the offset of the start of the invalid UTF-8 character is placed in
2138    the first output vector element (\fIovector[0]\fP) and a reason code is placed
2139    in the second element (\fIovector[1]\fP). The reason codes are given names in
2140    the \fBpcre.h\fP header file:
2141    .sp
2142      PCRE_UTF8_ERR1
2143      PCRE_UTF8_ERR2
2144      PCRE_UTF8_ERR3
2145      PCRE_UTF8_ERR4
2146      PCRE_UTF8_ERR5
2147    .sp
2148    The string ends with a truncated UTF-8 character; the code specifies how many
2149    bytes are missing (1 to 5). Although RFC 3629 restricts UTF-8 characters to be
2150    no longer than 4 bytes, the encoding scheme (originally defined by RFC 2279)
2151    allows for up to 6 bytes, and this is checked first; hence the possibility of
2152    4 or 5 missing bytes.
2153    .sp
2154      PCRE_UTF8_ERR6
2155      PCRE_UTF8_ERR7
2156      PCRE_UTF8_ERR8
2157      PCRE_UTF8_ERR9
2158      PCRE_UTF8_ERR10
2159    .sp
2160    The two most significant bits of the 2nd, 3rd, 4th, 5th, or 6th byte of the
2161    character do not have the binary value 0b10 (that is, either the most
2162    significant bit is 0, or the next bit is 1).
2163    .sp
2164      PCRE_UTF8_ERR11
2165      PCRE_UTF8_ERR12
2166    .sp
2167    A character that is valid by the RFC 2279 rules is either 5 or 6 bytes long;
2168    these code points are excluded by RFC 3629.
2169    .sp
2170      PCRE_UTF8_ERR13
2171    .sp
2172    A 4-byte character has a value greater than 0x10fff; these code points are
2173    excluded by RFC 3629.
2174    .sp
2175      PCRE_UTF8_ERR14
2176    .sp
2177    A 3-byte character has a value in the range 0xd800 to 0xdfff; this range of
2178    code points are reserved by RFC 3629 for use with UTF-16, and so are excluded
2179    from UTF-8.
2180    .sp
2181      PCRE_UTF8_ERR15
2182      PCRE_UTF8_ERR16
2183      PCRE_UTF8_ERR17
2184      PCRE_UTF8_ERR18
2185      PCRE_UTF8_ERR19
2186    .sp
2187    A 2-, 3-, 4-, 5-, or 6-byte character is "overlong", that is, it codes for a
2188    value that can be represented by fewer bytes, which is invalid. For example,
2189    the two bytes 0xc0, 0xae give the value 0x2e, whose correct coding uses just
2190    one byte.
2191    .sp
2192      PCRE_UTF8_ERR20
2193    .sp
2194    The two most significant bits of the first byte of a character have the binary
2195    value 0b10 (that is, the most significant bit is 1 and the second is 0). Such a
2196    byte can only validly occur as the second or subsequent byte of a multi-byte
2197    character.
2198    .sp
2199      PCRE_UTF8_ERR21
2200    .sp
2201    The first byte of a character has the value 0xfe or 0xff. These values can
2202    never occur in a valid UTF-8 string.
2203  .  .
2204  .  .
2205  .SH "EXTRACTING CAPTURED SUBSTRINGS BY NUMBER"  .SH "EXTRACTING CAPTURED SUBSTRINGS BY NUMBER"
# Line 1966  names are not included in the compiled c Line 2364  names are not included in the compiled c
2364  numbers. For this reason, the use of different names for subpatterns of the  numbers. For this reason, the use of different names for subpatterns of the
2365  same number causes an error at compile time.  same number causes an error at compile time.
2366  .  .
2367    .
2368  .SH "DUPLICATE SUBPATTERN NAMES"  .SH "DUPLICATE SUBPATTERN NAMES"
2369  .rs  .rs
2370  .sp  .sp
# Line 1999  fourth are pointers to variables which a Line 2398  fourth are pointers to variables which a
2398  has run, they point to the first and last entries in the name-to-number table  has run, they point to the first and last entries in the name-to-number table
2399  for the given name. The function itself returns the length of each entry, or  for the given name. The function itself returns the length of each entry, or
2400  PCRE_ERROR_NOSUBSTRING (-7) if there are none. The format of the table is  PCRE_ERROR_NOSUBSTRING (-7) if there are none. The format of the table is
2401  described above in the section entitled \fIInformation about a pattern\fP.  described above in the section entitled \fIInformation about a pattern\fP
2402    .\" HTML <a href="#infoaboutpattern">
2403    .\" </a>
2404    above.
2405    .\"
2406  Given all the relevant entries for the name, you can extract each of their  Given all the relevant entries for the name, you can extract each of their
2407  numbers, and hence the captured data, if any.  numbers, and hence the captured data, if any.
2408  .  .
# Line 2026  other alternatives. Ultimately, when it Line 2429  other alternatives. Ultimately, when it
2429  will yield PCRE_ERROR_NOMATCH.  will yield PCRE_ERROR_NOMATCH.
2430  .  .
2431  .  .
2432    .SH "OBTAINING AN ESTIMATE OF STACK USAGE"
2433    .rs
2434    .sp
2435    Matching certain patterns using \fBpcre_exec()\fP can use a lot of process
2436    stack, which in certain environments can be rather limited in size. Some users
2437    find it helpful to have an estimate of the amount of stack that is used by
2438    \fBpcre_exec()\fP, to help them set recursion limits, as described in the
2439    .\" HREF
2440    \fBpcrestack\fP
2441    .\"
2442    documentation. The estimate that is output by \fBpcretest\fP when called with
2443    the \fB-m\fP and \fB-C\fP options is obtained by calling \fBpcre_exec\fP with
2444    the values NULL, NULL, NULL, -999, and -999 for its first five arguments.
2445    .P
2446    Normally, if its first argument is NULL, \fBpcre_exec()\fP immediately returns
2447    the negative error code PCRE_ERROR_NULL, but with this special combination of
2448    arguments, it returns instead a negative number whose absolute value is the
2449    approximate stack frame size in bytes. (A negative number is used so that it is
2450    clear that no match has happened.) The value is approximate because in some
2451    cases, recursive calls to \fBpcre_exec()\fP occur when there are one or two
2452    additional variables on the stack.
2453    .P
2454    If PCRE has been compiled to use the heap instead of the stack for recursion,
2455    the value returned is the size of each block that is obtained from the heap.
2456    .
2457    .
2458  .\" HTML <a name="dfamatch"></a>  .\" HTML <a name="dfamatch"></a>
2459  .SH "MATCHING A PATTERN: THE ALTERNATIVE FUNCTION"  .SH "MATCHING A PATTERN: THE ALTERNATIVE FUNCTION"
2460  .rs  .rs
# Line 2129  match. There is more discussion of this Line 2558  match. There is more discussion of this
2558  .\"  .\"
2559  documentation.  documentation.
2560  .  .
2561    .
2562  .SS "Successful returns from \fBpcre_dfa_exec()\fP"  .SS "Successful returns from \fBpcre_dfa_exec()\fP"
2563  .rs  .rs
2564  .sp  .sp
# Line 2160  returns data, even though the meaning of Line 2590  returns data, even though the meaning of
2590  The strings are returned in reverse order of length; that is, the longest  The strings are returned in reverse order of length; that is, the longest
2591  matching string is given first. If there were too many matches to fit into  matching string is given first. If there were too many matches to fit into
2592  \fIovector\fP, the yield of the function is zero, and the vector is filled with  \fIovector\fP, the yield of the function is zero, and the vector is filled with
2593  the longest matches.  the longest matches. Unlike \fBpcre_exec()\fP, \fBpcre_dfa_exec()\fP can use
2594    the entire \fIovector\fP for returning matched strings.
2595    .
2596  .  .
2597  .SS "Error returns from \fBpcre_dfa_exec()\fP"  .SS "Error returns from \fBpcre_dfa_exec()\fP"
2598  .rs  .rs
# Line 2189  group. These are not supported. Line 2621  group. These are not supported.
2621    PCRE_ERROR_DFA_UMLIMIT    (-18)    PCRE_ERROR_DFA_UMLIMIT    (-18)
2622  .sp  .sp
2623  This return is given if \fBpcre_dfa_exec()\fP is called with an \fIextra\fP  This return is given if \fBpcre_dfa_exec()\fP is called with an \fIextra\fP
2624  block that contains a setting of the \fImatch_limit\fP field. This is not  block that contains a setting of the \fImatch_limit\fP or
2625  supported (it is meaningless).  \fImatch_limit_recursion\fP fields. This is not supported (these fields are
2626    meaningless for DFA matching).
2627  .sp  .sp
2628    PCRE_ERROR_DFA_WSSIZE     (-19)    PCRE_ERROR_DFA_WSSIZE     (-19)
2629  .sp  .sp
# Line 2203  When a recursive subpattern is processed Line 2636  When a recursive subpattern is processed
2636  recursively, using private vectors for \fIovector\fP and \fIworkspace\fP. This  recursively, using private vectors for \fIovector\fP and \fIworkspace\fP. This
2637  error is given if the output vector is not large enough. This should be  error is given if the output vector is not large enough. This should be
2638  extremely rare, as a vector of size 1000 is used.  extremely rare, as a vector of size 1000 is used.
2639    .sp
2640      PCRE_ERROR_DFA_BADRESTART (-30)
2641    .sp
2642    When \fBpcre_dfa_exec()\fP is called with the \fBPCRE_DFA_RESTART\fP option,
2643    some plausibility checks are made on the contents of the workspace, which
2644    should contain data about the previous partial match. If any of these checks
2645    fail, this error is given.
2646  .  .
2647  .  .
2648  .SH "SEE ALSO"  .SH "SEE ALSO"
2649  .rs  .rs
2650  .sp  .sp
2651  \fBpcrebuild\fP(3), \fBpcrecallout\fP(3), \fBpcrecpp(3)\fP(3),  \fBpcre16\fP(3), \fBpcrebuild\fP(3), \fBpcrecallout\fP(3), \fBpcrecpp(3)\fP(3),
2652  \fBpcrematching\fP(3), \fBpcrepartial\fP(3), \fBpcreposix\fP(3),  \fBpcrematching\fP(3), \fBpcrepartial\fP(3), \fBpcreposix\fP(3),
2653  \fBpcreprecompile\fP(3), \fBpcresample\fP(3), \fBpcrestack\fP(3).  \fBpcreprecompile\fP(3), \fBpcresample\fP(3), \fBpcrestack\fP(3).
2654  .  .
# Line 2227  Cambridge CB2 3QH, England. Line 2667  Cambridge CB2 3QH, England.
2667  .rs  .rs
2668  .sp  .sp
2669  .nf  .nf
2670  Last updated: 06 November 2010  Last updated: 17 June 2012
2671  Copyright (c) 1997-2010 University of Cambridge.  Copyright (c) 1997-2012 University of Cambridge.
2672  .fi  .fi

Legend:
Removed from v.568  
changed lines
  Added in v.978

  ViewVC Help
Powered by ViewVC 1.1.5