/[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 461 by ph10, Mon Oct 5 10:59:35 2009 UTC revision 788 by ph10, Tue Dec 6 15:38:01 2011 UTC
# Line 1  Line 1 
1  .TH PCREAPI 3  .TH PCREAPI 3
2  .SH NAME  .SH NAME
3  PCRE - Perl-compatible regular expressions  PCRE - Perl-compatible regular expressions
4  .SH "PCRE NATIVE API"  .SH "PCRE NATIVE API BASIC FUNCTIONS"
5  .rs  .rs
6  .sp  .sp
7  .B #include <pcre.h>  .B #include <pcre.h>
# Line 25  PCRE - Perl-compatible regular expressio Line 25  PCRE - Perl-compatible regular expressio
25  .ti +5n  .ti +5n
26  .B const char **\fIerrptr\fP);  .B const char **\fIerrptr\fP);
27  .PP  .PP
28    .B void pcre_free_study(pcre_extra *\fIextra\fP);
29    .PP
30  .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,"
31  .ti +5n  .ti +5n
32  .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,  .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
33  .ti +5n  .ti +5n
34  .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP);  .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP);
35    .
36    .
37    .SH "PCRE NATIVE API AUXILIARY FUNCTIONS"
38    .rs
39    .sp
40    .B pcre_jit_stack *pcre_jit_stack_alloc(int \fIstartsize\fP, int \fImaxsize\fP);
41    .PP
42    .B void pcre_jit_stack_free(pcre_jit_stack *\fIstack\fP);
43    .PP
44    .B void pcre_assign_jit_stack(pcre_extra *\fIextra\fP,
45    .ti +5n
46    .B pcre_jit_callback \fIcallback\fP, void *\fIdata\fP);
47  .PP  .PP
48  .B int pcre_dfa_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"  .B int pcre_dfa_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
49  .ti +5n  .ti +5n
# Line 97  PCRE - Perl-compatible regular expressio Line 111  PCRE - Perl-compatible regular expressio
111  .B int pcre_config(int \fIwhat\fP, void *\fIwhere\fP);  .B int pcre_config(int \fIwhat\fP, void *\fIwhere\fP);
112  .PP  .PP
113  .B char *pcre_version(void);  .B char *pcre_version(void);
114  .PP  .
115    .
116    .SH "PCRE NATIVE API INDIRECTED FUNCTIONS"
117    .rs
118    .sp
119  .B void *(*pcre_malloc)(size_t);  .B void *(*pcre_malloc)(size_t);
120  .PP  .PP
121  .B void (*pcre_free)(void *);  .B void (*pcre_free)(void *);
# Line 114  PCRE - Perl-compatible regular expressio Line 132  PCRE - Perl-compatible regular expressio
132  .sp  .sp
133  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
134  also some wrapper functions that correspond to the POSIX regular expression  also some wrapper functions that correspond to the POSIX regular expression
135  API. These are described in the  API, but they do not give access to all the functionality. They are described
136    in the
137  .\" HREF  .\" HREF
138  \fBpcreposix\fP  \fBpcreposix\fP
139  .\"  .\"
140  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++
141  wrapper is distributed with PCRE. It is documented in the  wrapper is also distributed with PCRE. It is documented in the
142  .\" HREF  .\" HREF
143  \fBpcrecpp\fP  \fBpcrecpp\fP
144  .\"  .\"
# Line 132  an application that uses PCRE. The heade Line 151  an application that uses PCRE. The heade
151  and PCRE_MINOR to contain the major and minor release numbers for the library.  and PCRE_MINOR to contain the major and minor release numbers for the library.
152  Applications can use these to include support for different releases of PCRE.  Applications can use these to include support for different releases of PCRE.
153  .P  .P
154    In a Windows environment, if you want to statically link an application program
155    against a non-dll \fBpcre.a\fP file, you must define PCRE_STATIC before
156    including \fBpcre.h\fP or \fBpcrecpp.h\fP, because otherwise the
157    \fBpcre_malloc()\fP and \fBpcre_free()\fP exported functions will be declared
158    \fB__declspec(dllimport)\fP, with unwanted results.
159    .P
160  The functions \fBpcre_compile()\fP, \fBpcre_compile2()\fP, \fBpcre_study()\fP,  The functions \fBpcre_compile()\fP, \fBpcre_compile2()\fP, \fBpcre_study()\fP,
161  and \fBpcre_exec()\fP are used for compiling and matching regular expressions  and \fBpcre_exec()\fP are used for compiling and matching regular expressions
162  in a Perl-compatible manner. A sample program that demonstrates the simplest  in a Perl-compatible manner. A sample program that demonstrates the simplest
# Line 146  documentation, and the Line 171  documentation, and the
171  .\"  .\"
172  documentation describes how to compile and run it.  documentation describes how to compile and run it.
173  .P  .P
174    Just-in-time compiler support is an optional feature of PCRE that can be built
175    in appropriate hardware environments. It greatly speeds up the matching
176    performance of many patterns. Simple programs can easily request that it be
177    used if available, by setting an option that is ignored when it is not
178    relevant. More complicated programs might need to make use of the functions
179    \fBpcre_jit_stack_alloc()\fP, \fBpcre_jit_stack_free()\fP, and
180    \fBpcre_assign_jit_stack()\fP in order to control the JIT code's memory usage.
181    These functions are discussed in the
182    .\" HREF
183    \fBpcrejit\fP
184    .\"
185    documentation.
186    .P
187  A second matching function, \fBpcre_dfa_exec()\fP, which is not  A second matching function, \fBpcre_dfa_exec()\fP, which is not
188  Perl-compatible, is also provided. This uses a different algorithm for the  Perl-compatible, is also provided. This uses a different algorithm for the
189  matching. The alternative algorithm finds all possible matches (at a given  matching. The alternative algorithm finds all possible matches (at a given
# Line 276  callout function pointed to by \fBpcre_c Line 314  callout function pointed to by \fBpcre_c
314  .P  .P
315  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
316  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.
317    .P
318    If the just-in-time optimization feature is being used, it needs separate
319    memory stack areas for each thread. See the
320    .\" HREF
321    \fBpcrejit\fP
322    .\"
323    documentation for more details.
324  .  .
325  .  .
326  .SH "SAVING PRECOMPILED PATTERNS FOR LATER USE"  .SH "SAVING PRECOMPILED PATTERNS FOR LATER USE"
# Line 318  otherwise it is set to zero. Line 363  otherwise it is set to zero.
363  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
364  properties is available; otherwise it is set to zero.  properties is available; otherwise it is set to zero.
365  .sp  .sp
366      PCRE_CONFIG_JIT
367    .sp
368    The output is an integer that is set to one if support for just-in-time
369    compiling is available; otherwise it is set to zero.
370    .sp
371    PCRE_CONFIG_NEWLINE    PCRE_CONFIG_NEWLINE
372  .sp  .sp
373  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 422  within the pattern (see the detailed des Line 472  within the pattern (see the detailed des
472  documentation). For those options that can be different in different parts of  documentation). For those options that can be different in different parts of
473  the pattern, the contents of the \fIoptions\fP argument specifies their  the pattern, the contents of the \fIoptions\fP argument specifies their
474  settings at the start of compilation and execution. The PCRE_ANCHORED,  settings at the start of compilation and execution. The PCRE_ANCHORED,
475  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
476  of matching as well as at compile time.  PCRE_NO_START_OPT options can be set at the time of matching as well as at
477    compile time.
478  .P  .P
479  If \fIerrptr\fP is NULL, \fBpcre_compile()\fP returns NULL immediately.  If \fIerrptr\fP is NULL, \fBpcre_compile()\fP returns NULL immediately.
480  Otherwise, if compilation of a pattern fails, \fBpcre_compile()\fP returns  Otherwise, if compilation of a pattern fails, \fBpcre_compile()\fP returns
481  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
482  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
483  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
484  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
485  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
486  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
487  carried out when the whole pattern has been scanned; in this case the offset is  that of the first byte of the failing character. Also, some errors are not
488  set to the end of the pattern.  detected until checks are carried out when the whole pattern has been scanned;
489    in these cases the offset passed back is the length of the pattern.
490    .P
491    Note that the offset is in bytes, not characters, even in UTF-8 mode. It may
492    sometimes point into the middle of a UTF-8 character.
493  .P  .P
494  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
495  \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 513  pattern. Line 568  pattern.
568  .sp  .sp
569    PCRE_DOTALL    PCRE_DOTALL
570  .sp  .sp
571  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
572  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
573  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,
574  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
575  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
576  the setting of this option.  (?s) option setting. A negative class such as [^a] always matches newline
577    characters, independent of the setting of this option.
578  .sp  .sp
579    PCRE_DUPNAMES    PCRE_DUPNAMES
580  .sp  .sp
# Line 540  unescaped # outside a character class an Line 596  unescaped # outside a character class an
596  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
597  pattern by a (?x) option setting.  pattern by a (?x) option setting.
598  .P  .P
599    Which characters are interpreted as newlines is controlled by the options
600    passed to \fBpcre_compile()\fP or by a special sequence at the start of the
601    pattern, as described in the section entitled
602    .\" HTML <a href="pcrepattern.html#newlines">
603    .\" </a>
604    "Newline conventions"
605    .\"
606    in the \fBpcrepattern\fP documentation. Note that the end of this type of
607    comment is a literal newline sequence in the pattern; escape sequences that
608    happen to represent a newline do not count.
609    .P
610  This option makes it possible to include comments inside complicated patterns.  This option makes it possible to include comments inside complicated patterns.
611  Note, however, that this applies only to data characters. Whitespace characters  Note, however, that this applies only to data characters. Whitespace characters
612  may never appear within special character sequences in a pattern, for example  may never appear within special character sequences in a pattern, for example
613  within the sequence (?( which introduces a conditional subpattern.  within the sequence (?( that introduces a conditional subpattern.
614  .sp  .sp
615    PCRE_EXTRA    PCRE_EXTRA
616  .sp  .sp
# Line 553  set, any backslash in a pattern that is Line 620  set, any backslash in a pattern that is
620  special meaning causes an error, thus reserving these combinations for future  special meaning causes an error, thus reserving these combinations for future
621  expansion. By default, as in Perl, a backslash followed by a letter with no  expansion. By default, as in Perl, a backslash followed by a letter with no
622  special meaning is treated as a literal. (Perl can, however, be persuaded to  special meaning is treated as a literal. (Perl can, however, be persuaded to
623  give a warning for this.) There are at present no other features controlled by  give an error for this, by running it with the -w option.) There are at present
624  this option. It can also be set by a (?X) option setting within a pattern.  no other features controlled by this option. It can also be set by a (?X)
625    option setting within a pattern.
626  .sp  .sp
627    PCRE_FIRSTLINE    PCRE_FIRSTLINE
628  .sp  .sp
# Line 575  character). Thus, the pattern AB]CD beco Line 643  character). Thus, the pattern AB]CD beco
643  string (by default this causes the current matching alternative to fail). A  string (by default this causes the current matching alternative to fail). A
644  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
645  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.
646    .P
647    (3) \eU matches an upper case "U" character; by default \eU causes a compile
648    time error (Perl uses \eU to upper case subsequent characters).
649    .P
650    (4) \eu matches a lower case "u" character unless it is followed by four
651    hexadecimal digits, in which case the hexadecimal number defines the code point
652    to match. By default, \eu causes a compile time error (Perl uses it to upper
653    case the following character).
654    .P
655    (5) \ex matches a lower case "x" character unless it is followed by two
656    hexadecimal digits, in which case the hexadecimal number defines the code point
657    to match. By default, as in Perl, a hexadecimal number is always expected after
658    \ex, but it may have zero, one, or two digits (so, for example, \exz matches a
659    binary zero character followed by z).
660  .sp  .sp
661    PCRE_MULTILINE    PCRE_MULTILINE
662  .sp  .sp
# Line 617  option, the combination may or may not b Line 699  option, the combination may or may not b
699  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
700  other combinations may yield unused numbers and cause an error.  other combinations may yield unused numbers and cause an error.
701  .P  .P
702  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
703  pattern is if PCRE_EXTENDED is set, and an unescaped # outside a character  compiling is when PCRE_EXTENDED is set. CR and LF are whitespace characters,
704  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
705  line break sequence. In other circumstances, line break sequences are treated  indicates a comment that lasts until after the next line break sequence. In
706  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
707  as whitespace characters and are therefore ignored.  data.
708  .P  .P
709  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
710  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 635  were followed by ?: but named parenthese Line 717  were followed by ?: but named parenthese
717  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
718  in Perl.  in Perl.
719  .sp  .sp
720      NO_START_OPTIMIZE
721    .sp
722    This is an option that acts at matching time; that is, it is really an option
723    for \fBpcre_exec()\fP or \fBpcre_dfa_exec()\fP. If it is set at compile time,
724    it is remembered with the compiled pattern and assumed at matching time. For
725    details see the discussion of PCRE_NO_START_OPTIMIZE
726    .\" HTML <a href="#execoptions">
727    .\" </a>
728    below.
729    .\"
730    .sp
731      PCRE_UCP
732    .sp
733    This option changes the way PCRE processes \eB, \eb, \eD, \ed, \eS, \es, \eW,
734    \ew, and some of the POSIX character classes. By default, only ASCII characters
735    are recognized, but if PCRE_UCP is set, Unicode properties are used instead to
736    classify characters. More details are given in the section on
737    .\" HTML <a href="pcre.html#genericchartypes">
738    .\" </a>
739    generic character types
740    .\"
741    in the
742    .\" HREF
743    \fBpcrepattern\fP
744    .\"
745    page. If you set PCRE_UCP, matching one of the items it affects takes much
746    longer. The option is available only if PCRE has been compiled with Unicode
747    property support.
748    .sp
749    PCRE_UNGREEDY    PCRE_UNGREEDY
750  .sp  .sp
751  This option inverts the "greediness" of the quantifiers so that they are not  This option inverts the "greediness" of the quantifiers so that they are not
# Line 648  of UTF-8 characters instead of single-by Line 759  of UTF-8 characters instead of single-by
759  available only when PCRE is built to include UTF-8 support. If not, the use  available only when PCRE is built to include UTF-8 support. If not, the use
760  of this option provokes an error. Details of how this option changes the  of this option provokes an error. Details of how this option changes the
761  behaviour of PCRE are given in the  behaviour of PCRE are given in the
 .\" HTML <a href="pcre.html#utf8support">  
 .\" </a>  
 section on UTF-8 support  
 .\"  
 in the main  
762  .\" HREF  .\" HREF
763  \fBpcre\fP  \fBpcreunicode\fP
764  .\"  .\"
765  page.  page.
766  .sp  .sp
# Line 724  out of use. To avoid confusion, they hav Line 830  out of use. To avoid confusion, they hav
830    34  character value in \ex{...} sequence is too large    34  character value in \ex{...} sequence is too large
831    35  invalid condition (?(0)    35  invalid condition (?(0)
832    36  \eC not allowed in lookbehind assertion    36  \eC not allowed in lookbehind assertion
833    37  PCRE does not support \eL, \el, \eN, \eU, or \eu    37  PCRE does not support \eL, \el, \eN{name}, \eU, or \eu
834    38  number after (?C is > 255    38  number after (?C is > 255
835    39  closing ) for (?C expected    39  closing ) for (?C expected
836    40  recursive call could loop indefinitely    40  recursive call could loop indefinitely
# Line 740  out of use. To avoid confusion, they hav Line 846  out of use. To avoid confusion, they hav
846    50  [this code is not in use]    50  [this code is not in use]
847    51  octal value is greater than \e377 (not in UTF-8 mode)    51  octal value is greater than \e377 (not in UTF-8 mode)
848    52  internal error: overran compiling workspace    52  internal error: overran compiling workspace
849    53  internal error: previously-checked referenced subpattern not found    53  internal error: previously-checked referenced subpattern
850            not found
851    54  DEFINE group contains more than one branch    54  DEFINE group contains more than one branch
852    55  repeating a DEFINE group is not allowed    55  repeating a DEFINE group is not allowed
853    56  inconsistent NEWLINE options    56  inconsistent NEWLINE options
854    57  \eg is not followed by a braced, angle-bracketed, or quoted    57  \eg is not followed by a braced, angle-bracketed, or quoted
855          name/number or by a plain number          name/number or by a plain number
856    58  a numbered reference must not be zero    58  a numbered reference must not be zero
857    59  (*VERB) with an argument is not supported    59  an argument is not allowed for (*ACCEPT), (*FAIL), or (*COMMIT)
858    60  (*VERB) not recognized    60  (*VERB) not recognized
859    61  number is too big    61  number is too big
860    62  subpattern name expected    62  subpattern name expected
861    63  digit expected after (?+    63  digit expected after (?+
862    64  ] is an invalid data character in JavaScript compatibility mode    64  ] is an invalid data character in JavaScript compatibility mode
863      65  different names for subpatterns of the same number are
864            not allowed
865      66  (*MARK) must have an argument
866      67  this version of PCRE is not compiled with PCRE_UCP support
867      68  \ec must be followed by an ASCII character
868      69  \ek is not followed by a braced, angle-bracketed, or quoted name
869  .sp  .sp
870  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
871  be used if the limits were changed when PCRE was built.  be used if the limits were changed when PCRE was built.
872  .  .
873  .  .
874    .\" HTML <a name="studyingapattern"></a>
875  .SH "STUDYING A PATTERN"  .SH "STUDYING A PATTERN"
876  .rs  .rs
877  .sp  .sp
# Line 788  If studying the pattern does not produce Line 902  If studying the pattern does not produce
902  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
903  \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.
904  .P  .P
905  The second argument of \fBpcre_study()\fP contains option bits. At present, no  The second argument of \fBpcre_study()\fP contains option bits. There is only
906  options are defined, and this argument should always be zero.  one option: PCRE_STUDY_JIT_COMPILE. If this is set, and the just-in-time
907    compiler is available, the pattern is further compiled into machine code that
908    executes much faster than the \fBpcre_exec()\fP matching function. If
909    the just-in-time compiler is not available, this option is ignored. All other
910    bits in the \fIoptions\fP argument must be zero.
911    .P
912    JIT compilation is a heavyweight optimization. It can take some time for
913    patterns to be analyzed, and for one-off matches and simple patterns the
914    benefit of faster execution might be offset by a much slower study time.
915    Not all patterns can be optimized by the JIT compiler. For those that cannot be
916    handled, matching automatically falls back to the \fBpcre_exec()\fP
917    interpreter. For more details, see the
918    .\" HREF
919    \fBpcrejit\fP
920    .\"
921    documentation.
922  .P  .P
923  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
924  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 798  static string that is part of the librar Line 927  static string that is part of the librar
927  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
928  sure that it has run successfully.  sure that it has run successfully.
929  .P  .P
930  This is a typical call to \fBpcre_study\fP():  When you are finished with a pattern, you can free the memory used for the
931    study data by calling \fBpcre_free_study()\fP. This function was added to the
932    API for release 8.20. For earlier versions, the memory could be freed with
933    \fBpcre_free()\fP, just like the pattern itself. This will still work in cases
934    where PCRE_STUDY_JIT_COMPILE is not used, but it is advisable to change to the
935    new function when convenient.
936    .P
937    This is a typical way in which \fBpcre_study\fP() is used (except that in a
938    real application there should be tests for errors):
939  .sp  .sp
940    pcre_extra *pe;    int rc;
941    pe = pcre_study(    pcre *re;
942      pcre_extra *sd;
943      re = pcre_compile("pattern", 0, &error, &erroroffset, NULL);
944      sd = pcre_study(
945      re,             /* result of pcre_compile() */      re,             /* result of pcre_compile() */
946      0,              /* no options exist */      0,              /* no options */
947      &error);        /* set to NULL or points to a message */      &error);        /* set to NULL or points to a message */
948      rc = pcre_exec(   /* see below for details of pcre_exec() options */
949        re, sd, "subject", 7, 0, 0, ovector, 30);
950      ...
951      pcre_free_study(sd);
952      pcre_free(re);
953  .sp  .sp
954  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
955  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 818  Studying a pattern is also useful for no Line 963  Studying a pattern is also useful for no
963  single fixed starting character. A bitmap of possible starting bytes is  single fixed starting character. A bitmap of possible starting bytes is
964  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
965  matching.  matching.
966    .P
967    These two optimizations apply to both \fBpcre_exec()\fP and
968    \fBpcre_dfa_exec()\fP. However, they are not used by \fBpcre_exec()\fP if
969    \fBpcre_study()\fP is called with the PCRE_STUDY_JIT_COMPILE option, and
970    just-in-time compiling is successful. The optimizations can be disabled by
971    setting the PCRE_NO_START_OPTIMIZE option when calling \fBpcre_exec()\fP or
972    \fBpcre_dfa_exec()\fP. You might want to do this if your pattern contains
973    callouts or (*MARK) (which cannot be handled by the JIT compiler), and you want
974    to make use of these facilities in cases where matching fails. See the
975    discussion of PCRE_NO_START_OPTIMIZE
976    .\" HTML <a href="#execoptions">
977    .\" </a>
978    below.
979    .\"
980  .  .
981  .  .
982  .\" HTML <a name="localesupport"></a>  .\" HTML <a name="localesupport"></a>
# Line 827  matching. Line 986  matching.
986  PCRE handles caseless matching, and determines whether characters are letters,  PCRE handles caseless matching, and determines whether characters are letters,
987  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
988  value. When running in UTF-8 mode, this applies only to characters with codes  value. When running in UTF-8 mode, this applies only to characters with codes
989  less than 128. Higher-valued codes never match escapes such as \ew or \ed, but  less than 128. By default, higher-valued codes never match escapes such as \ew
990  can be tested with \ep if PCRE is built with Unicode character property  or \ed, but they can be tested with \ep if PCRE is built with Unicode character
991  support. The use of locales with Unicode is discouraged. If you are handling  property support. Alternatively, the PCRE_UCP option can be set at compile
992  characters with codes greater than 128, you should either use UTF-8 and  time; this causes \ew and friends to use Unicode property support instead of
993  Unicode, or use locales, but not try to mix the two.  built-in tables. The use of locales with Unicode is discouraged. If you are
994    handling characters with codes greater than 128, you should either use UTF-8
995    and Unicode, or use locales, but not try to mix the two.
996  .P  .P
997  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
998  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 876  one in which it was compiled. Passing ta Line 1037  one in which it was compiled. Passing ta
1037  below in the section on matching a pattern.  below in the section on matching a pattern.
1038  .  .
1039  .  .
1040    .\" HTML <a name="infoaboutpattern"></a>
1041  .SH "INFORMATION ABOUT A PATTERN"  .SH "INFORMATION ABOUT A PATTERN"
1042  .rs  .rs
1043  .sp  .sp
# Line 907  check against passing an arbitrary memor Line 1069  check against passing an arbitrary memor
1069    size_t length;    size_t length;
1070    rc = pcre_fullinfo(    rc = pcre_fullinfo(
1071      re,               /* result of pcre_compile() */      re,               /* result of pcre_compile() */
1072      pe,               /* result of pcre_study(), or NULL */      sd,               /* result of pcre_study(), or NULL */
1073      PCRE_INFO_SIZE,   /* what is required */      PCRE_INFO_SIZE,   /* what is required */
1074      &length);         /* where to put the data */      &length);         /* where to put the data */
1075  .sp  .sp
# Line 972  Return 1 if the (?J) or (?-J) option set Line 1134  Return 1 if the (?J) or (?-J) option set
1134  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
1135  (?-J) set and unset the local PCRE_DUPNAMES option, respectively.  (?-J) set and unset the local PCRE_DUPNAMES option, respectively.
1136  .sp  .sp
1137      PCRE_INFO_JIT
1138    .sp
1139    Return 1 if the pattern was studied with the PCRE_STUDY_JIT_COMPILE option, and
1140    just-in-time compiling was successful. The fourth argument should point to an
1141    \fBint\fP variable. A return value of 0 means that JIT support is not available
1142    in this version of PCRE, or that the pattern was not studied with the
1143    PCRE_STUDY_JIT_COMPILE option, or that the JIT compiler could not handle this
1144    particular pattern. See the
1145    .\" HREF
1146    \fBpcrejit\fP
1147    .\"
1148    documentation for details of what can and cannot be handled.
1149    .sp
1150      PCRE_INFO_JITSIZE
1151    .sp
1152    If the pattern was successfully studied with the PCRE_STUDY_JIT_COMPILE option,
1153    return the size of the JIT compiled code, otherwise return zero. The fourth
1154    argument should point to a \fBsize_t\fP variable.
1155    .sp
1156    PCRE_INFO_LASTLITERAL    PCRE_INFO_LASTLITERAL
1157  .sp  .sp
1158  Return the value of the rightmost literal byte that must exist in any matched  Return the value of the rightmost literal byte that must exist in any matched
# Line 1087  For such patterns, the PCRE_ANCHORED bit Line 1268  For such patterns, the PCRE_ANCHORED bit
1268  .sp  .sp
1269    PCRE_INFO_SIZE    PCRE_INFO_SIZE
1270  .sp  .sp
1271  Return the size of the compiled pattern, that is, the value that was passed as  Return the size of the compiled pattern. The fourth argument should point to a
1272  the argument to \fBpcre_malloc()\fP when PCRE was getting memory in which to  \fBsize_t\fP variable. This value does not include the size of the \fBpcre\fP
1273  place the compiled data. The fourth argument should point to a \fBsize_t\fP  structure that is returned by \fBpcre_compile()\fP. The value that is passed as
1274  variable.  the argument to \fBpcre_malloc()\fP when \fBpcre_compile()\fP is getting memory
1275    in which to place the compiled data is the value returned by this option plus
1276    the size of the \fBpcre\fP structure. Studying a compiled pattern, with or
1277    without JIT, does not alter the value returned by this option.
1278  .sp  .sp
1279    PCRE_INFO_STUDYSIZE    PCRE_INFO_STUDYSIZE
1280  .sp  .sp
1281  Return the size of the data block pointed to by the \fIstudy_data\fP field in  Return the size of the data block pointed to by the \fIstudy_data\fP field in a
1282  a \fBpcre_extra\fP block. That is, it is the value that was passed to  \fBpcre_extra\fP block. If \fBpcre_extra\fP is NULL, or there is no study data,
1283  \fBpcre_malloc()\fP when PCRE was getting memory into which to place the data  zero is returned. The fourth argument should point to a \fBsize_t\fP variable.
1284  created by \fBpcre_study()\fP. If \fBpcre_extra\fP is NULL, or there is no  The \fIstudy_data\fP field is set by \fBpcre_study()\fP to record information
1285  study data, zero is returned. The fourth argument should point to a  that will speed up matching (see the section entitled
1286  \fBsize_t\fP variable.  .\" HTML <a href="#studyingapattern">
1287    .\" </a>
1288    "Studying a pattern"
1289    .\"
1290    above). The format of the \fIstudy_data\fP block is private, but its length
1291    is made available via this option so that it can be saved and restored (see the
1292    .\" HREF
1293    \fBpcreprecompile\fP
1294    .\"
1295    documentation for details).
1296  .  .
1297  .  .
1298  .SH "OBSOLETE INFO FUNCTION"  .SH "OBSOLETE INFO FUNCTION"
# Line 1161  is different. (This seems a highly unlik Line 1354  is different. (This seems a highly unlik
1354  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
1355  compiled pattern, which is passed in the \fIcode\fP argument. If the  compiled pattern, which is passed in the \fIcode\fP argument. If the
1356  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
1357  \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
1358  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
1359  also an alternative matching function, which is described  different subject strings with the same pattern.
1360    .P
1361    This function is the main matching facility of the library, and it operates in
1362    a Perl-like manner. For specialist use there is also an alternative matching
1363    function, which is described
1364  .\" HTML <a href="#dfamatch">  .\" HTML <a href="#dfamatch">
1365  .\" </a>  .\" </a>
1366  below  below
# Line 1194  Here is an example of a simple call to \ Line 1391  Here is an example of a simple call to \
1391      ovector,        /* vector of integers for substring information */      ovector,        /* vector of integers for substring information */
1392      30);            /* number of elements (NOT size in bytes) */      30);            /* number of elements (NOT size in bytes) */
1393  .  .
1394    .
1395  .\" HTML <a name="extradata"></a>  .\" HTML <a name="extradata"></a>
1396  .SS "Extra data for \fBpcre_exec()\fR"  .SS "Extra data for \fBpcre_exec()\fR"
1397  .rs  .rs
# Line 1206  fields (not necessarily in this order): Line 1404  fields (not necessarily in this order):
1404  .sp  .sp
1405    unsigned long int \fIflags\fP;    unsigned long int \fIflags\fP;
1406    void *\fIstudy_data\fP;    void *\fIstudy_data\fP;
1407      void *\fIexecutable_jit\fP;
1408    unsigned long int \fImatch_limit\fP;    unsigned long int \fImatch_limit\fP;
1409    unsigned long int \fImatch_limit_recursion\fP;    unsigned long int \fImatch_limit_recursion\fP;
1410    void *\fIcallout_data\fP;    void *\fIcallout_data\fP;
1411    const unsigned char *\fItables\fP;    const unsigned char *\fItables\fP;
1412      unsigned char **\fImark\fP;
1413  .sp  .sp
1414  The \fIflags\fP field is a bitmap that specifies which of the other fields  The \fIflags\fP field is a bitmap that specifies which of the other fields
1415  are set. The flag bits are:  are set. The flag bits are:
1416  .sp  .sp
1417    PCRE_EXTRA_STUDY_DATA    PCRE_EXTRA_STUDY_DATA
1418      PCRE_EXTRA_EXECUTABLE_JIT
1419    PCRE_EXTRA_MATCH_LIMIT    PCRE_EXTRA_MATCH_LIMIT
1420    PCRE_EXTRA_MATCH_LIMIT_RECURSION    PCRE_EXTRA_MATCH_LIMIT_RECURSION
1421    PCRE_EXTRA_CALLOUT_DATA    PCRE_EXTRA_CALLOUT_DATA
1422    PCRE_EXTRA_TABLES    PCRE_EXTRA_TABLES
1423      PCRE_EXTRA_MARK
1424  .sp  .sp
1425  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
1426  \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
1427  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
1428  the block by setting the other fields and their corresponding flag bits.  should not set these yourself, but you may add to the block by setting the
1429    other fields and their corresponding flag bits.
1430  .P  .P
1431  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
1432  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,
1433  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
1434  classic example is a pattern that uses nested unlimited repeats.  classic example is a pattern that uses nested unlimited repeats.
1435  .P  .P
1436  Internally, PCRE uses a function called \fBmatch()\fP which it calls repeatedly  Internally, \fBpcre_exec()\fP uses a function called \fBmatch()\fP, which it
1437  (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
1438  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
1439  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
1440  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
1441  string.  in the subject string.
1442    .P
1443    When \fBpcre_exec()\fP is called with a pattern that was successfully studied
1444    with the PCRE_STUDY_JIT_COMPILE option, the way that the matching is executed
1445    is entirely different. However, there is still the possibility of runaway
1446    matching that goes on for a very long time, and so the \fImatch_limit\fP value
1447    is also used in this case (but in a different way) to limit how long the
1448    matching can continue.
1449  .P  .P
1450  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
1451  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 1250  limits the depth of recursion. The recur Line 1460  limits the depth of recursion. The recur
1460  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.
1461  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.
1462  .P  .P
1463  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
1464  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
1465  amount of heap memory that can be used.  stack, the amount of heap memory that can be used. This limit is not relevant,
1466    and is ignored, if the pattern was successfully studied with
1467    PCRE_STUDY_JIT_COMPILE.
1468  .P  .P
1469  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
1470  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 1281  called. See the Line 1493  called. See the
1493  \fBpcreprecompile\fP  \fBpcreprecompile\fP
1494  .\"  .\"
1495  documentation for a discussion of saving compiled patterns for later use.  documentation for a discussion of saving compiled patterns for later use.
1496    .P
1497    If PCRE_EXTRA_MARK is set in the \fIflags\fP field, the \fImark\fP field must
1498    be set to point to a \fBchar *\fP variable. If the pattern contains any
1499    backtracking control verbs such as (*MARK:NAME), and the execution ends up with
1500    a name to pass back, a pointer to the name string (zero terminated) is placed
1501    in the variable pointed to by the \fImark\fP field. The names are within the
1502    compiled pattern; if you wish to retain such a name you must copy it before
1503    freeing the memory of a compiled pattern. If there is no name to pass back, the
1504    variable pointed to by the \fImark\fP field set to NULL. For details of the
1505    backtracking control verbs, see the section entitled
1506    .\" HTML <a href="pcrepattern#backtrackcontrol">
1507    .\" </a>
1508    "Backtracking control"
1509    .\"
1510    in the
1511    .\" HREF
1512    \fBpcrepattern\fP
1513    .\"
1514    documentation.
1515    .
1516  .  .
1517  .\" HTML <a name="execoptions"></a>  .\" HTML <a name="execoptions"></a>
1518  .SS "Option bits for \fBpcre_exec()\fP"  .SS "Option bits for \fBpcre_exec()\fP"
# Line 1291  zero. The only bits that may be set are Line 1523  zero. The only bits that may be set are
1523  PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NOTEMPTY_ATSTART,  PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NOTEMPTY_ATSTART,
1524  PCRE_NO_START_OPTIMIZE, PCRE_NO_UTF8_CHECK, PCRE_PARTIAL_SOFT, and  PCRE_NO_START_OPTIMIZE, PCRE_NO_UTF8_CHECK, PCRE_PARTIAL_SOFT, and
1525  PCRE_PARTIAL_HARD.  PCRE_PARTIAL_HARD.
1526    .P
1527    If the pattern was successfully studied with the PCRE_STUDY_JIT_COMPILE option,
1528    the only supported options for JIT execution are PCRE_NO_UTF8_CHECK,
1529    PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, and PCRE_NOTEMPTY_ATSTART. Note in
1530    particular that partial matching is not supported. If an unsupported option is
1531    used, JIT execution is disabled and the normal interpretive code in
1532    \fBpcre_exec()\fP is run.
1533  .sp  .sp
1534    PCRE_ANCHORED    PCRE_ANCHORED
1535  .sp  .sp
# Line 1386  the Line 1625  the
1625  .\" HREF  .\" HREF
1626  \fBpcredemo\fP  \fBpcredemo\fP
1627  .\"  .\"
1628  sample program.  sample program. In the most general case, you have to check to see if the
1629    newline convention recognizes CRLF as a newline, and if so, and the current
1630    character is CR followed by LF, advance the starting offset by two characters
1631    instead of one.
1632  .sp  .sp
1633    PCRE_NO_START_OPTIMIZE    PCRE_NO_START_OPTIMIZE
1634  .sp  .sp
1635  There are a number of optimizations that \fBpcre_exec()\fP uses at the start of  There are a number of optimizations that \fBpcre_exec()\fP uses at the start of
1636  a match, in order to speed up the process. For example, if it is known that a  a match, in order to speed up the process. For example, if it is known that an
1637  match must start with a specific character, it searches the subject for that  unanchored match must start with a specific character, it searches the subject
1638  character, and fails immediately if it cannot find it, without actually running  for that character, and fails immediately if it cannot find it, without
1639  the main matching function. When callouts are in use, these optimizations can  actually running the main matching function. This means that a special item
1640  cause them to be skipped. This option disables the "start-up" optimizations,  such as (*COMMIT) at the start of a pattern is not considered until after a
1641  causing performance to suffer, but ensuring that the callouts do occur.  suitable starting point for the match has been found. When callouts or (*MARK)
1642    items are in use, these "start-up" optimizations can cause them to be skipped
1643    if the pattern is never actually used. The start-up optimizations are in effect
1644    a pre-scan of the subject that takes place before the pattern is run.
1645    .P
1646    The PCRE_NO_START_OPTIMIZE option disables the start-up optimizations, possibly
1647    causing performance to suffer, but ensuring that in cases where the result is
1648    "no match", the callouts do occur, and that items such as (*COMMIT) and (*MARK)
1649    are considered at every possible starting position in the subject string. If
1650    PCRE_NO_START_OPTIMIZE is set at compile time, it cannot be unset at matching
1651    time.
1652    .P
1653    Setting PCRE_NO_START_OPTIMIZE can change the outcome of a matching operation.
1654    Consider the pattern
1655    .sp
1656      (*COMMIT)ABC
1657    .sp
1658    When this is compiled, PCRE records the fact that a match must start with the
1659    character "A". Suppose the subject string is "DEFABC". The start-up
1660    optimization scans along the subject, finds "A" and runs the first match
1661    attempt from there. The (*COMMIT) item means that the pattern must match the
1662    current starting position, which in this case, it does. However, if the same
1663    match is run with PCRE_NO_START_OPTIMIZE set, the initial scan along the
1664    subject string does not happen. The first match attempt is run starting from
1665    "D" and when this fails, (*COMMIT) prevents any further matches being tried, so
1666    the overall result is "no match". If the pattern is studied, more start-up
1667    optimizations may be used. For example, a minimum length for the subject may be
1668    recorded. Consider the pattern
1669    .sp
1670      (*MARK:A)(X|Y)
1671    .sp
1672    The minimum length for a match is one character. If the subject is "ABC", there
1673    will be attempts to match "ABC", "BC", "C", and then finally an empty string.
1674    If the pattern is studied, the final attempt does not take place, because PCRE
1675    knows that the subject is too short, and so the (*MARK) is never encountered.
1676    In this case, studying the pattern does not affect the overall match result,
1677    which is still "no match", but it does affect the auxiliary information that is
1678    returned.
1679  .sp  .sp
1680    PCRE_NO_UTF8_CHECK    PCRE_NO_UTF8_CHECK
1681  .sp  .sp
# Line 1414  in the main Line 1693  in the main
1693  \fBpcre\fP  \fBpcre\fP
1694  .\"  .\"
1695  page. If an invalid UTF-8 sequence of bytes is found, \fBpcre_exec()\fP returns  page. If an invalid UTF-8 sequence of bytes is found, \fBpcre_exec()\fP returns
1696  the error PCRE_ERROR_BADUTF8. If \fIstartoffset\fP contains an invalid value,  the error PCRE_ERROR_BADUTF8 or, if PCRE_PARTIAL_HARD is set and the problem is
1697  PCRE_ERROR_BADUTF8_OFFSET is returned.  a truncated UTF-8 character at the end of the subject, PCRE_ERROR_SHORTUTF8. In
1698    both cases, information about the precise nature of the error may also be
1699    returned (see the descriptions of these errors in the section entitled \fIError
1700    return values from\fP \fBpcre_exec()\fP
1701    .\" HTML <a href="#errorlist">
1702    .\" </a>
1703    below).
1704    .\"
1705    If \fIstartoffset\fP contains a value that does not point to the start of a
1706    UTF-8 character (or to the end of the subject), PCRE_ERROR_BADUTF8_OFFSET is
1707    returned.
1708  .P  .P
1709  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
1710  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
1711  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
1712  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
1713  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
1714  the value of \fIstartoffset\fP points to the start of a UTF-8 character. When  the value of \fIstartoffset\fP points to the start of a UTF-8 character (or the
1715  PCRE_NO_UTF8_CHECK is set, the effect of passing an invalid UTF-8 string as a  end of the subject). When PCRE_NO_UTF8_CHECK is set, the effect of passing an
1716  subject, or a value of \fIstartoffset\fP that does not point to the start of a  invalid UTF-8 string as a subject or an invalid value of \fIstartoffset\fP is
1717  UTF-8 character, is undefined. Your program may crash.  undefined. Your program may crash.
1718  .sp  .sp
1719    PCRE_PARTIAL_HARD    PCRE_PARTIAL_HARD
1720    PCRE_PARTIAL_SOFT    PCRE_PARTIAL_SOFT
# Line 1434  These options turn on the partial matchi Line 1723  These options turn on the partial matchi
1723  compatibility, PCRE_PARTIAL is a synonym for PCRE_PARTIAL_SOFT. A partial match  compatibility, PCRE_PARTIAL is a synonym for PCRE_PARTIAL_SOFT. A partial match
1724  occurs if the end of the subject string is reached successfully, but there are  occurs if the end of the subject string is reached successfully, but there are
1725  not enough subject characters to complete the match. If this happens when  not enough subject characters to complete the match. If this happens when
1726  PCRE_PARTIAL_HARD is set, \fBpcre_exec()\fP immediately returns  PCRE_PARTIAL_SOFT (but not PCRE_PARTIAL_HARD) is set, matching continues by
1727  PCRE_ERROR_PARTIAL. Otherwise, if PCRE_PARTIAL_SOFT is set, matching continues  testing any remaining alternatives. Only if no complete match can be found is
1728  by testing any other alternatives. Only if they all fail is PCRE_ERROR_PARTIAL  PCRE_ERROR_PARTIAL returned instead of PCRE_ERROR_NOMATCH. In other words,
1729  returned (instead of PCRE_ERROR_NOMATCH). The portion of the string that  PCRE_PARTIAL_SOFT says that the caller is prepared to handle a partial match,
1730  was inspected when the partial match was found is set as the first matching  but only if no complete match can be found.
1731  string. There is a more detailed discussion in the  .P
1732    If PCRE_PARTIAL_HARD is set, it overrides PCRE_PARTIAL_SOFT. In this case, if a
1733    partial match is found, \fBpcre_exec()\fP immediately returns
1734    PCRE_ERROR_PARTIAL, without considering any other alternatives. In other words,
1735    when PCRE_PARTIAL_HARD is set, a partial match is considered to be more
1736    important that an alternative complete match.
1737    .P
1738    In both cases, the portion of the string that was inspected when the partial
1739    match was found is set as the first matching string. There is a more detailed
1740    discussion of partial and multi-segment matching, with examples, in the
1741  .\" HREF  .\" HREF
1742  \fBpcrepartial\fP  \fBpcrepartial\fP
1743  .\"  .\"
1744  documentation.  documentation.
1745  .  .
1746    .
1747  .SS "The string to be matched by \fBpcre_exec()\fP"  .SS "The string to be matched by \fBpcre_exec()\fP"
1748  .rs  .rs
1749  .sp  .sp
1750  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
1751  \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
1752  in \fIstartoffset\fP. In UTF-8 mode, the byte offset must point to the start of  in \fIstartoffset\fP. If this is negative or greater than the length of the
1753  a UTF-8 character. Unlike the pattern string, the subject may contain binary  subject, \fBpcre_exec()\fP returns PCRE_ERROR_BADOFFSET. When the starting
1754  zero bytes. When the starting offset is zero, the search for a match starts at  offset is zero, the search for a match starts at the beginning of the subject,
1755  the beginning of the subject, and this is by far the most common case.  and this is by far the most common case. In UTF-8 mode, the byte offset must
1756    point to the start of a UTF-8 character (or the end of the subject). Unlike the
1757    pattern string, the subject may contain binary zero bytes.
1758  .P  .P
1759  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
1760  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 1473  start of the subject, which is deemed to Line 1774  start of the subject, which is deemed to
1774  set to 4, it finds the second occurrence of "iss" because it is able to look  set to 4, it finds the second occurrence of "iss" because it is able to look
1775  behind the starting point to discover that it is preceded by a letter.  behind the starting point to discover that it is preceded by a letter.
1776  .P  .P
1777    Finding all the matches in a subject is tricky when the pattern can match an
1778    empty string. It is possible to emulate Perl's /g behaviour by first trying the
1779    match again at the same offset, with the PCRE_NOTEMPTY_ATSTART and
1780    PCRE_ANCHORED options, and then if that fails, advancing the starting offset
1781    and trying an ordinary match again. There is some code that demonstrates how to
1782    do this in the
1783    .\" HREF
1784    \fBpcredemo\fP
1785    .\"
1786    sample program. In the most general case, you have to check to see if the
1787    newline convention recognizes CRLF as a newline, and if so, and the current
1788    character is CR followed by LF, advance the starting offset by two characters
1789    instead of one.
1790    .P
1791  If a non-zero starting offset is passed when the pattern is anchored, one  If a non-zero starting offset is passed when the pattern is anchored, one
1792  attempt to match at the given offset is made. This can only succeed if the  attempt to match at the given offset is made. This can only succeed if the
1793  pattern does not require the match to be at the start of the subject.  pattern does not require the match to be at the start of the subject.
1794  .  .
1795    .
1796  .SS "How \fBpcre_exec()\fP returns captured substrings"  .SS "How \fBpcre_exec()\fP returns captured substrings"
1797  .rs  .rs
1798  .sp  .sp
# Line 1520  string that it matched that is returned. Line 1836  string that it matched that is returned.
1836  .P  .P
1837  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
1838  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
1839  returns a value of zero. If the substring offsets are not of interest,  returns a value of zero. If neither the actual string matched not any captured
1840  \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
1841  \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
1842  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
1843  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
1844  advisable to supply an \fIovector\fP.  is usually advisable to supply an \fIovector\fP of reasonable size.
1845    .P
1846    There are some cases where zero is returned (indicating vector overflow) when
1847    in fact the vector is exactly the right size for the final match. For example,
1848    consider the pattern
1849    .sp
1850      (a)(?:(b)c|bd)
1851    .sp
1852    If a vector of 6 elements (allowing for only 1 captured substring) is given
1853    with subject string "abd", \fBpcre_exec()\fP will try to set the second
1854    captured string, thereby recording a vector overflow, before failing to match
1855    "c" and backing up to try the second alternative. The zero return, however,
1856    does correctly indicate that the maximum number of slots (namely 2) have been
1857    filled. In similar cases where there is temporary overflow, but the final
1858    number of used slots is actually less than the maximum, a non-zero value is
1859    returned.
1860  .P  .P
1861  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
1862  subpatterns there are in a compiled pattern. The smallest size for  subpatterns there are in a compiled pattern. The smallest size for
# Line 1543  Offset values that correspond to unused Line 1874  Offset values that correspond to unused
1874  expression are also set to -1. For example, if the string "abc" is matched  expression are also set to -1. For example, if the string "abc" is matched
1875  against the pattern (abc)(x(yz)?)? subpatterns 2 and 3 are not matched. The  against the pattern (abc)(x(yz)?)? subpatterns 2 and 3 are not matched. The
1876  return from the function is 2, because the highest used capturing subpattern  return from the function is 2, because the highest used capturing subpattern
1877  number is 1. However, you can refer to the offsets for the second and third  number is 1, and the offsets for for the second and third capturing subpatterns
1878  capturing subpatterns if you wish (assuming the vector is large enough, of  (assuming the vector is large enough, of course) are set to -1.
1879  course).  .P
1880    \fBNote\fP: Elements in the first two-thirds of \fIovector\fP that do not
1881    correspond to capturing parentheses in the pattern are never changed. That is,
1882    if a pattern contains \fIn\fP capturing parentheses, no more than
1883    \fIovector[0]\fP to \fIovector[2n+1]\fP are set by \fBpcre_exec()\fP. The other
1884    elements (in the first two-thirds) retain whatever values they previously had.
1885  .P  .P
1886  Some convenience functions are provided for extracting the captured substrings  Some convenience functions are provided for extracting the captured substrings
1887  as separate strings. These are described below.  as separate strings. These are described below.
1888  .  .
1889    .
1890  .\" HTML <a name="errorlist"></a>  .\" HTML <a name="errorlist"></a>
1891  .SS "Error return values from \fBpcre_exec()\fP"  .SS "Error return values from \fBpcre_exec()\fP"
1892  .rs  .rs
# Line 1591  If a pattern contains back references, b Line 1928  If a pattern contains back references, b
1928  gets a block of memory at the start of matching to use for this purpose. If the  gets a block of memory at the start of matching to use for this purpose. If the
1929  call via \fBpcre_malloc()\fP fails, this error is given. The memory is  call via \fBpcre_malloc()\fP fails, this error is given. The memory is
1930  automatically freed at the end of matching.  automatically freed at the end of matching.
1931    .P
1932    This error is also given if \fBpcre_stack_malloc()\fP fails in
1933    \fBpcre_exec()\fP. This can happen only when PCRE has been compiled with
1934    \fB--disable-stack-for-recursion\fP.
1935  .sp  .sp
1936    PCRE_ERROR_NOSUBSTRING    (-7)    PCRE_ERROR_NOSUBSTRING    (-7)
1937  .sp  .sp
# Line 1615  documentation for details. Line 1956  documentation for details.
1956  .sp  .sp
1957    PCRE_ERROR_BADUTF8        (-10)    PCRE_ERROR_BADUTF8        (-10)
1958  .sp  .sp
1959  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,
1960    and the PCRE_NO_UTF8_CHECK option was not set. If the size of the output vector
1961    (\fIovecsize\fP) is at least 2, the byte offset to the start of the the invalid
1962    UTF-8 character is placed in the first element, and a reason code is placed in
1963    the second element. The reason codes are listed in the
1964    .\" HTML <a href="#badutf8reasons">
1965    .\" </a>
1966    following section.
1967    .\"
1968    For backward compatibility, if PCRE_PARTIAL_HARD is set and the problem is a
1969    truncated UTF-8 character at the end of the subject (reason codes 1 to 5),
1970    PCRE_ERROR_SHORTUTF8 is returned instead of PCRE_ERROR_BADUTF8.
1971  .sp  .sp
1972    PCRE_ERROR_BADUTF8_OFFSET (-11)    PCRE_ERROR_BADUTF8_OFFSET (-11)
1973  .sp  .sp
1974  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
1975  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
1976    \fIstartoffset\fP did not point to the beginning of a UTF-8 character or the
1977    end of the subject.
1978  .sp  .sp
1979    PCRE_ERROR_PARTIAL        (-12)    PCRE_ERROR_PARTIAL        (-12)
1980  .sp  .sp
# Line 1655  description above. Line 2009  description above.
2009    PCRE_ERROR_BADNEWLINE     (-23)    PCRE_ERROR_BADNEWLINE     (-23)
2010  .sp  .sp
2011  An invalid combination of PCRE_NEWLINE_\fIxxx\fP options was given.  An invalid combination of PCRE_NEWLINE_\fIxxx\fP options was given.
2012    .sp
2013      PCRE_ERROR_BADOFFSET      (-24)
2014    .sp
2015    The value of \fIstartoffset\fP was negative or greater than the length of the
2016    subject, that is, the value in \fIlength\fP.
2017    .sp
2018      PCRE_ERROR_SHORTUTF8      (-25)
2019    .sp
2020    This error is returned instead of PCRE_ERROR_BADUTF8 when the subject string
2021    ends with a truncated UTF-8 character and the PCRE_PARTIAL_HARD option is set.
2022    Information about the failure is returned as for PCRE_ERROR_BADUTF8. It is in
2023    fact sufficient to detect this case, but this special error code for
2024    PCRE_PARTIAL_HARD precedes the implementation of returned information; it is
2025    retained for backwards compatibility.
2026    .sp
2027      PCRE_ERROR_RECURSELOOP    (-26)
2028    .sp
2029    This error is returned when \fBpcre_exec()\fP detects a recursion loop within
2030    the pattern. Specifically, it means that either the whole pattern or a
2031    subpattern has been called recursively for the second time at the same position
2032    in the subject string. Some simple patterns that might do this are detected and
2033    faulted at compile time, but more complicated cases, in particular mutual
2034    recursions between two different subpatterns, cannot be detected until run
2035    time.
2036    .sp
2037      PCRE_ERROR_JIT_STACKLIMIT (-27)
2038    .sp
2039    This error is returned when a pattern that was successfully studied using the
2040    PCRE_STUDY_JIT_COMPILE option is being matched, but the memory available for
2041    the just-in-time processing stack is not large enough. See the
2042    .\" HREF
2043    \fBpcrejit\fP
2044    .\"
2045    documentation for more details.
2046  .P  .P
2047  Error numbers -16 to -20 and -22 are not used by \fBpcre_exec()\fP.  Error numbers -16 to -20 and -22 are not used by \fBpcre_exec()\fP.
2048  .  .
2049  .  .
2050    .\" HTML <a name="badutf8reasons"></a>
2051    .SS "Reason codes for invalid UTF-8 strings"
2052    .rs
2053    .sp
2054    When \fBpcre_exec()\fP returns either PCRE_ERROR_BADUTF8 or
2055    PCRE_ERROR_SHORTUTF8, and the size of the output vector (\fIovecsize\fP) is at
2056    least 2, the offset of the start of the invalid UTF-8 character is placed in
2057    the first output vector element (\fIovector[0]\fP) and a reason code is placed
2058    in the second element (\fIovector[1]\fP). The reason codes are given names in
2059    the \fBpcre.h\fP header file:
2060    .sp
2061      PCRE_UTF8_ERR1
2062      PCRE_UTF8_ERR2
2063      PCRE_UTF8_ERR3
2064      PCRE_UTF8_ERR4
2065      PCRE_UTF8_ERR5
2066    .sp
2067    The string ends with a truncated UTF-8 character; the code specifies how many
2068    bytes are missing (1 to 5). Although RFC 3629 restricts UTF-8 characters to be
2069    no longer than 4 bytes, the encoding scheme (originally defined by RFC 2279)
2070    allows for up to 6 bytes, and this is checked first; hence the possibility of
2071    4 or 5 missing bytes.
2072    .sp
2073      PCRE_UTF8_ERR6
2074      PCRE_UTF8_ERR7
2075      PCRE_UTF8_ERR8
2076      PCRE_UTF8_ERR9
2077      PCRE_UTF8_ERR10
2078    .sp
2079    The two most significant bits of the 2nd, 3rd, 4th, 5th, or 6th byte of the
2080    character do not have the binary value 0b10 (that is, either the most
2081    significant bit is 0, or the next bit is 1).
2082    .sp
2083      PCRE_UTF8_ERR11
2084      PCRE_UTF8_ERR12
2085    .sp
2086    A character that is valid by the RFC 2279 rules is either 5 or 6 bytes long;
2087    these code points are excluded by RFC 3629.
2088    .sp
2089      PCRE_UTF8_ERR13
2090    .sp
2091    A 4-byte character has a value greater than 0x10fff; these code points are
2092    excluded by RFC 3629.
2093    .sp
2094      PCRE_UTF8_ERR14
2095    .sp
2096    A 3-byte character has a value in the range 0xd800 to 0xdfff; this range of
2097    code points are reserved by RFC 3629 for use with UTF-16, and so are excluded
2098    from UTF-8.
2099    .sp
2100      PCRE_UTF8_ERR15
2101      PCRE_UTF8_ERR16
2102      PCRE_UTF8_ERR17
2103      PCRE_UTF8_ERR18
2104      PCRE_UTF8_ERR19
2105    .sp
2106    A 2-, 3-, 4-, 5-, or 6-byte character is "overlong", that is, it codes for a
2107    value that can be represented by fewer bytes, which is invalid. For example,
2108    the two bytes 0xc0, 0xae give the value 0x2e, whose correct coding uses just
2109    one byte.
2110    .sp
2111      PCRE_UTF8_ERR20
2112    .sp
2113    The two most significant bits of the first byte of a character have the binary
2114    value 0b10 (that is, the most significant bit is 1 and the second is 0). Such a
2115    byte can only validly occur as the second or subsequent byte of a multi-byte
2116    character.
2117    .sp
2118      PCRE_UTF8_ERR21
2119    .sp
2120    The first byte of a character has the value 0xfe or 0xff. These values can
2121    never occur in a valid UTF-8 string.
2122    .
2123    .
2124  .SH "EXTRACTING CAPTURED SUBSTRINGS BY NUMBER"  .SH "EXTRACTING CAPTURED SUBSTRINGS BY NUMBER"
2125  .rs  .rs
2126  .sp  .sp
# Line 1821  names are not included in the compiled c Line 2283  names are not included in the compiled c
2283  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
2284  same number causes an error at compile time.  same number causes an error at compile time.
2285  .  .
2286    .
2287  .SH "DUPLICATE SUBPATTERN NAMES"  .SH "DUPLICATE SUBPATTERN NAMES"
2288  .rs  .rs
2289  .sp  .sp
# Line 1854  fourth are pointers to variables which a Line 2317  fourth are pointers to variables which a
2317  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
2318  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
2319  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
2320  described above in the section entitled \fIInformation about a pattern\fP.  described above in the section entitled \fIInformation about a pattern\fP
2321    .\" HTML <a href="#infoaboutpattern">
2322    .\" </a>
2323    above.
2324    .\"
2325  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
2326  numbers, and hence the captured data, if any.  numbers, and hence the captured data, if any.
2327  .  .
# Line 1939  Here is an example of a simple call to \ Line 2406  Here is an example of a simple call to \
2406  The unused bits of the \fIoptions\fP argument for \fBpcre_dfa_exec()\fP must be  The unused bits of the \fIoptions\fP argument for \fBpcre_dfa_exec()\fP must be
2407  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,
2408  PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NOTEMPTY_ATSTART,  PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NOTEMPTY_ATSTART,
2409  PCRE_NO_UTF8_CHECK, PCRE_PARTIAL_HARD, PCRE_PARTIAL_SOFT, PCRE_DFA_SHORTEST,  PCRE_NO_UTF8_CHECK, PCRE_BSR_ANYCRLF, PCRE_BSR_UNICODE, PCRE_NO_START_OPTIMIZE,
2410  and PCRE_DFA_RESTART. All but the last four of these are exactly the same as  PCRE_PARTIAL_HARD, PCRE_PARTIAL_SOFT, PCRE_DFA_SHORTEST, and PCRE_DFA_RESTART.
2411  for \fBpcre_exec()\fP, so their description is not repeated here.  All but the last four of these are exactly the same as for \fBpcre_exec()\fP,
2412    so their description is not repeated here.
2413  .sp  .sp
2414    PCRE_PARTIAL_HARD    PCRE_PARTIAL_HARD
2415    PCRE_PARTIAL_SOFT    PCRE_PARTIAL_SOFT
# Line 1956  is converted into PCRE_ERROR_PARTIAL if Line 2424  is converted into PCRE_ERROR_PARTIAL if
2424  there have been no complete matches, but there is still at least one matching  there have been no complete matches, but there is still at least one matching
2425  possibility. The portion of the string that was inspected when the longest  possibility. The portion of the string that was inspected when the longest
2426  partial match was found is set as the first matching string in both cases.  partial match was found is set as the first matching string in both cases.
2427    There is a more detailed discussion of partial and multi-segment matching, with
2428    examples, in the
2429    .\" HREF
2430    \fBpcrepartial\fP
2431    .\"
2432    documentation.
2433  .sp  .sp
2434    PCRE_DFA_SHORTEST    PCRE_DFA_SHORTEST
2435  .sp  .sp
# Line 1977  match. There is more discussion of this Line 2451  match. There is more discussion of this
2451  .\"  .\"
2452  documentation.  documentation.
2453  .  .
2454    .
2455  .SS "Successful returns from \fBpcre_dfa_exec()\fP"  .SS "Successful returns from \fBpcre_dfa_exec()\fP"
2456  .rs  .rs
2457  .sp  .sp
# Line 2008  returns data, even though the meaning of Line 2483  returns data, even though the meaning of
2483  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
2484  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
2485  \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
2486  the longest matches.  the longest matches. Unlike \fBpcre_exec()\fP, \fBpcre_dfa_exec()\fP can use
2487    the entire \fIovector\fP for returning matched strings.
2488    .
2489  .  .
2490  .SS "Error returns from \fBpcre_dfa_exec()\fP"  .SS "Error returns from \fBpcre_dfa_exec()\fP"
2491  .rs  .rs
# Line 2037  group. These are not supported. Line 2514  group. These are not supported.
2514    PCRE_ERROR_DFA_UMLIMIT    (-18)    PCRE_ERROR_DFA_UMLIMIT    (-18)
2515  .sp  .sp
2516  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
2517  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
2518  supported (it is meaningless).  \fImatch_limit_recursion\fP fields. This is not supported (these fields are
2519    meaningless for DFA matching).
2520  .sp  .sp
2521    PCRE_ERROR_DFA_WSSIZE     (-19)    PCRE_ERROR_DFA_WSSIZE     (-19)
2522  .sp  .sp
# Line 2075  Cambridge CB2 3QH, England. Line 2553  Cambridge CB2 3QH, England.
2553  .rs  .rs
2554  .sp  .sp
2555  .nf  .nf
2556  Last updated: 03 October 2009  Last updated: 02 December 2011
2557  Copyright (c) 1997-2009 University of Cambridge.  Copyright (c) 1997-2011 University of Cambridge.
2558  .fi  .fi

Legend:
Removed from v.461  
changed lines
  Added in v.788

  ViewVC Help
Powered by ViewVC 1.1.5