/[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 75 by nigel, Sat Feb 24 21:40:37 2007 UTC revision 93 by nigel, Sat Feb 24 21:41:42 2007 UTC
# Line 1  Line 1 
1  .TH PCRE 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"
# Line 15  PCRE - Perl-compatible regular expressio Line 15  PCRE - Perl-compatible regular expressio
15  .B const unsigned char *\fItableptr\fP);  .B const unsigned char *\fItableptr\fP);
16  .PP  .PP
17  .br  .br
18    .B pcre *pcre_compile2(const char *\fIpattern\fP, int \fIoptions\fP,
19    .ti +5n
20    .B int *\fIerrorcodeptr\fP,
21    .ti +5n
22    .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
23    .ti +5n
24    .B const unsigned char *\fItableptr\fP);
25    .PP
26    .br
27  .B pcre_extra *pcre_study(const pcre *\fIcode\fP, int \fIoptions\fP,  .B pcre_extra *pcre_study(const pcre *\fIcode\fP, int \fIoptions\fP,
28  .ti +5n  .ti +5n
29  .B const char **\fIerrptr\fP);  .B const char **\fIerrptr\fP);
# Line 27  PCRE - Perl-compatible regular expressio Line 36  PCRE - Perl-compatible regular expressio
36  .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP);  .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP);
37  .PP  .PP
38  .br  .br
39    .B int pcre_dfa_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
40    .ti +5n
41    .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
42    .ti +5n
43    .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP,
44    .ti +5n
45    .B int *\fIworkspace\fP, int \fIwscount\fP);
46    .PP
47    .br
48  .B int pcre_copy_named_substring(const pcre *\fIcode\fP,  .B int pcre_copy_named_substring(const pcre *\fIcode\fP,
49  .ti +5n  .ti +5n
50  .B const char *\fIsubject\fP, int *\fIovector\fP,  .B const char *\fIsubject\fP, int *\fIovector\fP,
# Line 57  PCRE - Perl-compatible regular expressio Line 75  PCRE - Perl-compatible regular expressio
75  .B const char *\fIname\fP);  .B const char *\fIname\fP);
76  .PP  .PP
77  .br  .br
78    .B int pcre_get_stringtable_entries(const pcre *\fIcode\fP,
79    .ti +5n
80    .B const char *\fIname\fP, char **\fIfirst\fP, char **\fIlast\fP);
81    .PP
82    .br
83  .B int pcre_get_substring(const char *\fIsubject\fP, int *\fIovector\fP,  .B int pcre_get_substring(const char *\fIsubject\fP, int *\fIovector\fP,
84  .ti +5n  .ti +5n
85  .B int \fIstringcount\fP, int \fIstringnumber\fP,  .B int \fIstringcount\fP, int \fIstringnumber\fP,
# Line 87  PCRE - Perl-compatible regular expressio Line 110  PCRE - Perl-compatible regular expressio
110  .B *\fIfirstcharptr\fP);  .B *\fIfirstcharptr\fP);
111  .PP  .PP
112  .br  .br
113    .B int pcre_refcount(pcre *\fIcode\fP, int \fIadjust\fP);
114    .PP
115    .br
116  .B int pcre_config(int \fIwhat\fP, void *\fIwhere\fP);  .B int pcre_config(int \fIwhat\fP, void *\fIwhere\fP);
117  .PP  .PP
118  .br  .br
# Line 111  PCRE - Perl-compatible regular expressio Line 137  PCRE - Perl-compatible regular expressio
137  .SH "PCRE API OVERVIEW"  .SH "PCRE API OVERVIEW"
138  .rs  .rs
139  .sp  .sp
140  PCRE has its own native API, which is described in this document. There is also  PCRE has its own native API, which is described in this document. There are
141  a set of wrapper functions that correspond to the POSIX regular expression API.  also some wrapper functions that correspond to the POSIX regular expression
142  These are described in the  API. These are described in the
143  .\" HREF  .\" HREF
144  \fBpcreposix\fP  \fBpcreposix\fP
145  .\"  .\"
146  documentation.  documentation. Both of these APIs define a set of C function calls. A C++
147    wrapper is distributed with PCRE. It is documented in the
148    .\" HREF
149    \fBpcrecpp\fP
150    .\"
151    page.
152  .P  .P
153  The native API function prototypes are defined in the header file \fBpcre.h\fP,  The native API C function prototypes are defined in the header file
154  and on Unix systems the library itself is called \fBlibpcre\fP. It can  \fBpcre.h\fP, and on Unix systems the library itself is called \fBlibpcre\fP.
155  normally be accessed by adding \fB-lpcre\fP to the command for linking an  It can normally be accessed by adding \fB-lpcre\fP to the command for linking
156  application that uses PCRE. The header file defines the macros PCRE_MAJOR and  an application that uses PCRE. The header file defines the macros PCRE_MAJOR
157  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.
158  Applications can use these to include support for different releases of PCRE.  Applications can use these to include support for different releases of PCRE.
159  .P  .P
160  The functions \fBpcre_compile()\fP, \fBpcre_study()\fP, and \fBpcre_exec()\fP  The functions \fBpcre_compile()\fP, \fBpcre_compile2()\fP, \fBpcre_study()\fP,
161  are used for compiling and matching regular expressions. A sample program that  and \fBpcre_exec()\fP are used for compiling and matching regular expressions
162  demonstrates the simplest way of using them is provided in the file called  in a Perl-compatible manner. A sample program that demonstrates the simplest
163  \fIpcredemo.c\fP in the source distribution. The  way of using them is provided in the file called \fIpcredemo.c\fP in the source
164    distribution. The
165  .\" HREF  .\" HREF
166  \fBpcresample\fP  \fBpcresample\fP
167  .\"  .\"
168  documentation describes how to run it.  documentation describes how to run it.
169  .P  .P
170    A second matching function, \fBpcre_dfa_exec()\fP, which is not
171    Perl-compatible, is also provided. This uses a different algorithm for the
172    matching. The alternative algorithm finds all possible matches (at a given
173    point in the subject), and scans the subject just once. However, this algorithm
174    does not return captured substrings. A description of the two matching
175    algorithms and their advantages and disadvantages is given in the
176    .\" HREF
177    \fBpcrematching\fP
178    .\"
179    documentation.
180    .P
181  In addition to the main compiling and matching functions, there are convenience  In addition to the main compiling and matching functions, there are convenience
182  functions for extracting captured substrings from a matched subject string.  functions for extracting captured substrings from a subject string that is
183  They are:  matched by \fBpcre_exec()\fP. They are:
184  .sp  .sp
185    \fBpcre_copy_substring()\fP    \fBpcre_copy_substring()\fP
186    \fBpcre_copy_named_substring()\fP    \fBpcre_copy_named_substring()\fP
# Line 145  They are: Line 188  They are:
188    \fBpcre_get_named_substring()\fP    \fBpcre_get_named_substring()\fP
189    \fBpcre_get_substring_list()\fP    \fBpcre_get_substring_list()\fP
190    \fBpcre_get_stringnumber()\fP    \fBpcre_get_stringnumber()\fP
191      \fBpcre_get_stringtable_entries()\fP
192  .sp  .sp
193  \fBpcre_free_substring()\fP and \fBpcre_free_substring_list()\fP are also  \fBpcre_free_substring()\fP and \fBpcre_free_substring_list()\fP are also
194  provided, to free the memory used for extracted strings.  provided, to free the memory used for extracted strings.
195  .P  .P
196  The function \fBpcre_maketables()\fP is used to build a set of character tables  The function \fBpcre_maketables()\fP is used to build a set of character tables
197  in the current locale for passing to \fBpcre_compile()\fP or \fBpcre_exec()\fP.  in the current locale for passing to \fBpcre_compile()\fP, \fBpcre_exec()\fP,
198  This is an optional facility that is provided for specialist use. Most  or \fBpcre_dfa_exec()\fP. This is an optional facility that is provided for
199  commonly, no special tables are passed, in which case internal tables that are  specialist use. Most commonly, no special tables are passed, in which case
200  generated when PCRE is built are used.  internal tables that are generated when PCRE is built are used.
201  .P  .P
202  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
203  compiled pattern; \fBpcre_info()\fP is an obsolete version that returns only  compiled pattern; \fBpcre_info()\fP is an obsolete version that returns only
# Line 161  some of the available information, but i Line 205  some of the available information, but i
205  The function \fBpcre_version()\fP returns a pointer to a string containing the  The function \fBpcre_version()\fP returns a pointer to a string containing the
206  version of PCRE and its date of release.  version of PCRE and its date of release.
207  .P  .P
208    The function \fBpcre_refcount()\fP maintains a reference count in a data block
209    containing a compiled pattern. This is provided for the benefit of
210    object-oriented applications.
211    .P
212  The global variables \fBpcre_malloc\fP and \fBpcre_free\fP initially contain  The global variables \fBpcre_malloc\fP and \fBpcre_free\fP initially contain
213  the entry points of the standard \fBmalloc()\fP and \fBfree()\fP functions,  the entry points of the standard \fBmalloc()\fP and \fBfree()\fP functions,
214  respectively. PCRE calls the memory management functions via these variables,  respectively. PCRE calls the memory management functions via these variables,
# Line 170  should be done before calling any PCRE f Line 218  should be done before calling any PCRE f
218  The global variables \fBpcre_stack_malloc\fP and \fBpcre_stack_free\fP are also  The global variables \fBpcre_stack_malloc\fP and \fBpcre_stack_free\fP are also
219  indirections to memory management functions. These special functions are used  indirections to memory management functions. These special functions are used
220  only when PCRE is compiled to use the heap for remembering data, instead of  only when PCRE is compiled to use the heap for remembering data, instead of
221  recursive function calls. This is a non-standard way of building PCRE, for use  recursive function calls, when running the \fBpcre_exec()\fP function. See the
222  in environments that have limited stacks. Because of the greater use of memory  .\" HREF
223  management, it runs more slowly. Separate functions are provided so that  \fBpcrebuild\fP
224  special-purpose external code can be used for this case. When used, these  .\"
225  functions are always called in a stack-like manner (last obtained, first  documentation for details of how to do this. It is a non-standard way of
226  freed), and always for memory blocks of the same size.  building PCRE, for use in environments that have limited stacks. Because of the
227    greater use of memory management, it runs more slowly. Separate functions are
228    provided so that special-purpose external code can be used for this case. When
229    used, these functions are always called in a stack-like manner (last obtained,
230    first freed), and always for memory blocks of the same size. There is a
231    discussion about PCRE's stack usage in the
232    .\" HREF
233    \fBpcrestack\fP
234    .\"
235    documentation.
236  .P  .P
237  The global variable \fBpcre_callout\fP initially contains NULL. It can be set  The global variable \fBpcre_callout\fP initially contains NULL. It can be set
238  by the caller to a "callout" function, which PCRE will then call at specified  by the caller to a "callout" function, which PCRE will then call at specified
# Line 186  points during a matching operation. Deta Line 243  points during a matching operation. Deta
243  documentation.  documentation.
244  .  .
245  .  .
246    .SH NEWLINES
247    .rs
248    .sp
249    PCRE supports four different conventions for indicating line breaks in
250    strings: a single CR (carriage return) character, a single LF (linefeed)
251    character, the two-character sequence CRLF, or any Unicode newline sequence.
252    The Unicode newline sequences are the three just mentioned, plus the single
253    characters VT (vertical tab, U+000B), FF (formfeed, U+000C), NEL (next line,
254    U+0085), LS (line separator, U+2028), and PS (paragraph separator, U+2029).
255    .P
256    Each of the first three conventions is used by at least one operating system as
257    its standard newline sequence. When PCRE is built, a default can be specified.
258    The default default is LF, which is the Unix standard. When PCRE is run, the
259    default can be overridden, either when a pattern is compiled, or when it is
260    matched.
261    .P
262    In the PCRE documentation the word "newline" is used to mean "the character or
263    pair of characters that indicate a line break". The choice of newline
264    convention affects the handling of the dot, circumflex, and dollar
265    metacharacters, the handling of #-comments in /x mode, and, when CRLF is a
266    recognized line ending sequence, the match position advancement for a
267    non-anchored pattern. The choice of newline convention does not affect the
268    interpretation of the \en or \er escape sequences.
269    .
270    .
271  .SH MULTITHREADING  .SH MULTITHREADING
272  .rs  .rs
273  .sp  .sp
# Line 238  properties is available; otherwise it is Line 320  properties is available; otherwise it is
320  .sp  .sp
321    PCRE_CONFIG_NEWLINE    PCRE_CONFIG_NEWLINE
322  .sp  .sp
323  The output is an integer that is set to the value of the code that is used for  The output is an integer whose value specifies the default character sequence
324  the newline character. It is either linefeed (10) or carriage return (13), and  that is recognized as meaning "newline". The four values that are supported
325  should normally be the standard character for your operating system.  are: 10 for LF, 13 for CR, 3338 for CRLF, and -1 for ANY. The default should
326    normally be the standard sequence for your operating system.
327  .sp  .sp
328    PCRE_CONFIG_LINK_SIZE    PCRE_CONFIG_LINK_SIZE
329  .sp  .sp
# Line 266  The output is an integer that gives the Line 349  The output is an integer that gives the
349  internal matching function calls in a \fBpcre_exec()\fP execution. Further  internal matching function calls in a \fBpcre_exec()\fP execution. Further
350  details are given with \fBpcre_exec()\fP below.  details are given with \fBpcre_exec()\fP below.
351  .sp  .sp
352      PCRE_CONFIG_MATCH_LIMIT_RECURSION
353    .sp
354    The output is an integer that gives the default limit for the depth of
355    recursion when calling the internal matching function in a \fBpcre_exec()\fP
356    execution. Further details are given with \fBpcre_exec()\fP below.
357    .sp
358    PCRE_CONFIG_STACKRECURSE    PCRE_CONFIG_STACKRECURSE
359  .sp  .sp
360  The output is an integer that is set to one if internal recursion is  The output is an integer that is set to one if internal recursion when running
361  implemented by recursive function calls that use the stack to remember their  \fBpcre_exec()\fP is implemented by recursive function calls that use the stack
362  state. This is the usual way that PCRE is compiled. The output is zero if PCRE  to remember their state. This is the usual way that PCRE is compiled. The
363  was compiled to use blocks of data on the heap instead of recursive function  output is zero if PCRE was compiled to use blocks of data on the heap instead
364  calls. In this case, \fBpcre_stack_malloc\fP and \fBpcre_stack_free\fP are  of recursive function calls. In this case, \fBpcre_stack_malloc\fP and
365  called to manage memory blocks on the heap, thus avoiding the use of the stack.  \fBpcre_stack_free\fP are called to manage memory blocks on the heap, thus
366    avoiding the use of the stack.
367  .  .
368  .  .
369  .SH "COMPILING A PATTERN"  .SH "COMPILING A PATTERN"
# Line 284  called to manage memory blocks on the he Line 374  called to manage memory blocks on the he
374  .B const char **\fIerrptr\fP, int *\fIerroffset\fP,  .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
375  .ti +5n  .ti +5n
376  .B const unsigned char *\fItableptr\fP);  .B const unsigned char *\fItableptr\fP);
377    .sp
378    .B pcre *pcre_compile2(const char *\fIpattern\fP, int \fIoptions\fP,
379    .ti +5n
380    .B int *\fIerrorcodeptr\fP,
381    .ti +5n
382    .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
383    .ti +5n
384    .B const unsigned char *\fItableptr\fP);
385  .P  .P
386  The function \fBpcre_compile()\fP is called to compile a pattern into an  Either of the functions \fBpcre_compile()\fP or \fBpcre_compile2()\fP can be
387  internal form. The pattern is a C string terminated by a binary zero, and  called to compile a pattern into an internal form. The only difference between
388  is passed in the \fIpattern\fP argument. A pointer to a single block of memory  the two interfaces is that \fBpcre_compile2()\fP has an additional argument,
389  that is obtained via \fBpcre_malloc\fP is returned. This contains the compiled  \fIerrorcodeptr\fP, via which a numerical error code can be returned.
390  code and related data. The \fBpcre\fP type is defined for the returned block;  .P
391  this is a typedef for a structure whose contents are not externally defined. It  The pattern is a C string terminated by a binary zero, and is passed in the
392  is up to the caller to free the memory when it is no longer required.  \fIpattern\fP argument. A pointer to a single block of memory that is obtained
393    via \fBpcre_malloc\fP is returned. This contains the compiled code and related
394    data. The \fBpcre\fP type is defined for the returned block; this is a typedef
395    for a structure whose contents are not externally defined. It is up to the
396    caller to free the memory (via \fBpcre_free\fP) when it is no longer required.
397  .P  .P
398  Although the compiled code of a PCRE regex is relocatable, that is, it does not  Although the compiled code of a PCRE regex is relocatable, that is, it does not
399  depend on memory location, the complete \fBpcre\fP data block is not  depend on memory location, the complete \fBpcre\fP data block is not
400  fully relocatable, because it may contain a copy of the \fItableptr\fP  fully relocatable, because it may contain a copy of the \fItableptr\fP
401  argument, which is an address (see below).  argument, which is an address (see below).
402  .P  .P
403  The \fIoptions\fP argument contains independent bits that affect the  The \fIoptions\fP argument contains various bit settings that affect the
404  compilation. It should be zero if no options are required. The available  compilation. It should be zero if no options are required. The available
405  options are described below. Some of them, in particular, those that are  options are described below. Some of them, in particular, those that are
406  compatible with Perl, can also be set and unset from within the pattern (see  compatible with Perl, can also be set and unset from within the pattern (see
# Line 308  the detailed description in the Line 410  the detailed description in the
410  .\"  .\"
411  documentation). For these options, the contents of the \fIoptions\fP argument  documentation). For these options, the contents of the \fIoptions\fP argument
412  specifies their initial settings at the start of compilation and execution. The  specifies their initial settings at the start of compilation and execution. The
413  PCRE_ANCHORED option can be set at the time of matching as well as at compile  PCRE_ANCHORED and PCRE_NEWLINE_\fIxxx\fP options can be set at the time of
414  time.  matching as well as at compile time.
415  .P  .P
416  If \fIerrptr\fP is NULL, \fBpcre_compile()\fP returns NULL immediately.  If \fIerrptr\fP is NULL, \fBpcre_compile()\fP returns NULL immediately.
417  Otherwise, if compilation of a pattern fails, \fBpcre_compile()\fP returns  Otherwise, if compilation of a pattern fails, \fBpcre_compile()\fP returns
418  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
419  error message. The offset from the start of the pattern to the character where  error message. This is a static string that is part of the library. You must
420  the error was discovered is placed in the variable pointed to by  not try to free it. The offset from the start of the pattern to the character
421    where the error was discovered is placed in the variable pointed to by
422  \fIerroffset\fP, which must not be NULL. If it is, an immediate error is given.  \fIerroffset\fP, which must not be NULL. If it is, an immediate error is given.
423  .P  .P
424    If \fBpcre_compile2()\fP is used instead of \fBpcre_compile()\fP, and the
425    \fIerrorcodeptr\fP argument is not NULL, a non-zero error code number is
426    returned via this argument in the event of an error. This is in addition to the
427    textual error message. Error codes and messages are listed below.
428    .P
429  If the final argument, \fItableptr\fP, is NULL, PCRE uses a default set of  If the final argument, \fItableptr\fP, is NULL, PCRE uses a default set of
430  character tables that are built when PCRE is compiled, using the default C  character tables that are built when PCRE is compiled, using the default C
431  locale. Otherwise, \fItableptr\fP must be an address that is the result of a  locale. Otherwise, \fItableptr\fP must be an address that is the result of a
# Line 362  documentation. Line 470  documentation.
470  .sp  .sp
471  If this bit is set, letters in the pattern match both upper and lower case  If this bit is set, letters in the pattern match both upper and lower case
472  letters. It is equivalent to Perl's /i option, and it can be changed within a  letters. It is equivalent to Perl's /i option, and it can be changed within a
473  pattern by a (?i) option setting. When running in UTF-8 mode, case support for  pattern by a (?i) option setting. In UTF-8 mode, PCRE always understands the
474  high-valued characters is available only when PCRE is built with Unicode  concept of case for characters whose values are less than 128, so caseless
475  character property support.  matching is always possible. For characters with higher values, the concept of
476    case is supported if PCRE is compiled with Unicode property support, but not
477    otherwise. If you want to use caseless matching for characters 128 and above,
478    you must ensure that PCRE is compiled with Unicode property support as well as
479    with UTF-8 support.
480  .sp  .sp
481    PCRE_DOLLAR_ENDONLY    PCRE_DOLLAR_ENDONLY
482  .sp  .sp
483  If this bit is set, a dollar metacharacter in the pattern matches only at the  If this bit is set, a dollar metacharacter in the pattern matches only at the
484  end of the subject string. Without this option, a dollar also matches  end of the subject string. Without this option, a dollar also matches
485  immediately before the final character if it is a newline (but not before any  immediately before a newline at the end of the string (but not before any other
486  other newlines). The PCRE_DOLLAR_ENDONLY option is ignored if PCRE_MULTILINE is  newlines). The PCRE_DOLLAR_ENDONLY option is ignored if PCRE_MULTILINE is set.
487  set. There is no equivalent to this option in Perl, and no way to set it within  There is no equivalent to this option in Perl, and no way to set it within a
488  a pattern.  pattern.
489  .sp  .sp
490    PCRE_DOTALL    PCRE_DOTALL
491  .sp  .sp
492  If this bit is set, a dot metacharater in the pattern matches all characters,  If this bit is set, a dot metacharater in the pattern matches all characters,
493  including newlines. Without it, newlines are excluded. This option is  including those that indicate newline. Without it, a dot does not match when
494  equivalent to Perl's /s option, and it can be changed within a pattern by a  the current position is at a newline. This option is equivalent to Perl's /s
495  (?s) option setting. A negative class such as [^a] always matches a newline  option, and it can be changed within a pattern by a (?s) option setting. A
496  character, independent of the setting of this option.  negative class such as [^a] always matches newline characters, independent of
497    the setting of this option.
498    .sp
499      PCRE_DUPNAMES
500    .sp
501    If this bit is set, names used to identify capturing subpatterns need not be
502    unique. This can be helpful for certain types of pattern when it is known that
503    only one instance of the named subpattern can ever be matched. There are more
504    details of named subpatterns below; see also the
505    .\" HREF
506    \fBpcrepattern\fP
507    .\"
508    documentation.
509  .sp  .sp
510    PCRE_EXTENDED    PCRE_EXTENDED
511  .sp  .sp
512  If this bit is set, whitespace data characters in the pattern are totally  If this bit is set, whitespace data characters in the pattern are totally
513  ignored except when escaped or inside a character class. Whitespace does not  ignored except when escaped or inside a character class. Whitespace does not
514  include the VT character (code 11). In addition, characters between an  include the VT character (code 11). In addition, characters between an
515  unescaped # outside a character class and the next newline character,  unescaped # outside a character class and the next newline, inclusive, are also
516  inclusive, are also ignored. This is equivalent to Perl's /x option, and it can  ignored. This is equivalent to Perl's /x option, and it can be changed within a
517  be changed within a pattern by a (?x) option setting.  pattern by a (?x) option setting.
518  .P  .P
519  This option makes it possible to include comments inside complicated patterns.  This option makes it possible to include comments inside complicated patterns.
520  Note, however, that this applies only to data characters. Whitespace characters  Note, however, that this applies only to data characters. Whitespace characters
# Line 404  that is incompatible with Perl, but it i Line 528  that is incompatible with Perl, but it i
528  set, any backslash in a pattern that is followed by a letter that has no  set, any backslash in a pattern that is followed by a letter that has no
529  special meaning causes an error, thus reserving these combinations for future  special meaning causes an error, thus reserving these combinations for future
530  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
531  special meaning is treated as a literal. There are at present no other features  special meaning is treated as a literal. (Perl can, however, be persuaded to
532  controlled by this option. It can also be set by a (?X) option setting within a  give a warning for this.) There are at present no other features controlled by
533  pattern.  this option. It can also be set by a (?X) option setting within a pattern.
534    .sp
535      PCRE_FIRSTLINE
536    .sp
537    If this option is set, an unanchored pattern is required to match before or at
538    the first newline in the subject string, though the matched text may continue
539    over the newline.
540  .sp  .sp
541    PCRE_MULTILINE    PCRE_MULTILINE
542  .sp  .sp
# Line 418  terminating newline (unless PCRE_DOLLAR_ Line 548  terminating newline (unless PCRE_DOLLAR_
548  Perl.  Perl.
549  .P  .P
550  When PCRE_MULTILINE it is set, the "start of line" and "end of line" constructs  When PCRE_MULTILINE it is set, the "start of line" and "end of line" constructs
551  match immediately following or immediately before any newline in the subject  match immediately following or immediately before internal newlines in the
552  string, respectively, as well as at the very start and end. This is equivalent  subject string, respectively, as well as at the very start and end. This is
553  to Perl's /m option, and it can be changed within a pattern by a (?m) option  equivalent to Perl's /m option, and it can be changed within a pattern by a
554  setting. If there are no "\en" characters in a subject string, or no  (?m) option setting. If there are no newlines in a subject string, or no
555  occurrences of ^ or $ in a pattern, setting PCRE_MULTILINE has no effect.  occurrences of ^ or $ in a pattern, setting PCRE_MULTILINE has no effect.
556  .sp  .sp
557      PCRE_NEWLINE_CR
558      PCRE_NEWLINE_LF
559      PCRE_NEWLINE_CRLF
560      PCRE_NEWLINE_ANY
561    .sp
562    These options override the default newline definition that was chosen when PCRE
563    was built. Setting the first or the second specifies that a newline is
564    indicated by a single character (CR or LF, respectively). Setting
565    PCRE_NEWLINE_CRLF specifies that a newline is indicated by the two-character
566    CRLF sequence. Setting PCRE_NEWLINE_ANY specifies that any Unicode newline
567    sequence should be recognized. The Unicode newline sequences are the three just
568    mentioned, plus the single characters VT (vertical tab, U+000B), FF (formfeed,
569    U+000C), NEL (next line, U+0085), LS (line separator, U+2028), and PS
570    (paragraph separator, U+2029). The last two are recognized only in UTF-8 mode.
571    .P
572    The newline setting in the options word uses three bits that are treated
573    as a number, giving eight possibilities. Currently only five are used (default
574    plus the four values above). This means that if you set more than one newline
575    option, the combination may or may not be sensible. For example,
576    PCRE_NEWLINE_CR with PCRE_NEWLINE_LF is equivalent to PCRE_NEWLINE_CRLF, but
577    other combinations yield unused numbers and cause an error.
578    .P
579    The only time that a line break is specially recognized when compiling a
580    pattern is if PCRE_EXTENDED is set, and an unescaped # outside a character
581    class is encountered. This indicates a comment that lasts until after the next
582    line break sequence. In other circumstances, line break sequences are treated
583    as literal data, except that in PCRE_EXTENDED mode, both CR and LF are treated
584    as whitespace characters and are therefore ignored.
585    .P
586    The newline option that is set at compile time becomes the default that is used
587    for \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP, but it can be overridden.
588    .sp
589    PCRE_NO_AUTO_CAPTURE    PCRE_NO_AUTO_CAPTURE
590  .sp  .sp
591  If this option is set, it disables the use of numbered capturing parentheses in  If this option is set, it disables the use of numbered capturing parentheses in
# Line 463  automatically checked. If an invalid UTF Line 625  automatically checked. If an invalid UTF
625  valid, and you want to skip this check for performance reasons, you can set the  valid, and you want to skip this check for performance reasons, you can set the
626  PCRE_NO_UTF8_CHECK option. When it is set, the effect of passing an invalid  PCRE_NO_UTF8_CHECK option. When it is set, the effect of passing an invalid
627  UTF-8 string as a pattern is undefined. It may cause your program to crash.  UTF-8 string as a pattern is undefined. It may cause your program to crash.
628  Note that this option can also be passed to \fBpcre_exec()\fP, to suppress the  Note that this option can also be passed to \fBpcre_exec()\fP and
629  UTF-8 validity checking of subject strings.  \fBpcre_dfa_exec()\fP, to suppress the UTF-8 validity checking of subject
630    strings.
631    .
632    .
633    .SH "COMPILATION ERROR CODES"
634    .rs
635    .sp
636    The following table lists the error codes than may be returned by
637    \fBpcre_compile2()\fP, along with the error messages that may be returned by
638    both compiling functions. As PCRE has developed, some error codes have fallen
639    out of use. To avoid confusion, they have not been re-used.
640    .sp
641       0  no error
642       1  \e at end of pattern
643       2  \ec at end of pattern
644       3  unrecognized character follows \e
645       4  numbers out of order in {} quantifier
646       5  number too big in {} quantifier
647       6  missing terminating ] for character class
648       7  invalid escape sequence in character class
649       8  range out of order in character class
650       9  nothing to repeat
651      10  [this code is not in use]
652      11  internal error: unexpected repeat
653      12  unrecognized character after (?
654      13  POSIX named classes are supported only within a class
655      14  missing )
656      15  reference to non-existent subpattern
657      16  erroffset passed as NULL
658      17  unknown option bit(s) set
659      18  missing ) after comment
660      19  [this code is not in use]
661      20  regular expression too large
662      21  failed to get memory
663      22  unmatched parentheses
664      23  internal error: code overflow
665      24  unrecognized character after (?<
666      25  lookbehind assertion is not fixed length
667      26  malformed number or name after (?(
668      27  conditional group contains more than two branches
669      28  assertion expected after (?(
670      29  (?R or (?digits must be followed by )
671      30  unknown POSIX class name
672      31  POSIX collating elements are not supported
673      32  this version of PCRE is not compiled with PCRE_UTF8 support
674      33  [this code is not in use]
675      34  character value in \ex{...} sequence is too large
676      35  invalid condition (?(0)
677      36  \eC not allowed in lookbehind assertion
678      37  PCRE does not support \eL, \el, \eN, \eU, or \eu
679      38  number after (?C is > 255
680      39  closing ) for (?C expected
681      40  recursive call could loop indefinitely
682      41  unrecognized character after (?P
683      42  syntax error in subpattern name (missing terminator)
684      43  two named subpatterns have the same name
685      44  invalid UTF-8 string
686      45  support for \eP, \ep, and \eX has not been compiled
687      46  malformed \eP or \ep sequence
688      47  unknown property name after \eP or \ep
689      48  subpattern name is too long (maximum 32 characters)
690      49  too many named subpatterns (maximum 10,000)
691      50  repeated subpattern is too long
692      51  octal value is greater than \e377 (not in UTF-8 mode)
693      52  internal error: overran compiling workspace
694      53  internal error: previously-checked referenced subpattern not found
695      54  DEFINE group contains more than one branch
696      55  repeating a DEFINE group is not allowed
697      56  inconsistent NEWLINE options"
698  .  .
699  .  .
700  .SH "STUDYING A PATTERN"  .SH "STUDYING A PATTERN"
701  .rs  .rs
702  .sp  .sp
703  .B pcre_extra *pcre_study(const pcre *\fIcode\fP, int \fIoptions\fP,  .B pcre_extra *pcre_study(const pcre *\fIcode\fP, int \fIoptions\fP
704  .ti +5n  .ti +5n
705  .B const char **\fIerrptr\fP);  .B const char **\fIerrptr\fP);
706  .PP  .PP
# Line 492  below Line 722  below
722  .\"  .\"
723  in the section on matching a pattern.  in the section on matching a pattern.
724  .P  .P
725  If studying the pattern does not produce any additional information,  If studying the pattern does not produce any additional information
726  \fBpcre_study()\fP returns NULL. In that circumstance, if the calling program  \fBpcre_study()\fP returns NULL. In that circumstance, if the calling program
727  wants to pass any of the other fields to \fBpcre_exec()\fP, it must set up its  wants to pass any of the other fields to \fBpcre_exec()\fP, it must set up its
728  own \fBpcre_extra\fP block.  own \fBpcre_extra\fP block.
# Line 502  options are defined, and this argument s Line 732  options are defined, and this argument s
732  .P  .P
733  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
734  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
735  set to NULL. Otherwise it points to a textual error message. You should  set to NULL. Otherwise it is set to point to a textual error message. This is a
736  therefore test the error pointer for NULL after calling \fBpcre_study()\fP, to  static string that is part of the library. You must not try to free it. You
737  be sure that it has run successfully.  should test the error pointer for NULL after calling \fBpcre_study()\fP, to be
738    sure that it has run successfully.
739  .P  .P
740  This is a typical call to \fBpcre_study\fP():  This is a typical call to \fBpcre_study\fP():
741  .sp  .sp
# Line 523  bytes is created. Line 754  bytes is created.
754  .SH "LOCALE SUPPORT"  .SH "LOCALE SUPPORT"
755  .rs  .rs
756  .sp  .sp
757  PCRE handles caseless matching, and determines whether characters are letters,  PCRE handles caseless matching, and determines whether characters are letters
758  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
759  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
760  less than 128. Higher-valued codes never match escapes such as \ew or \ed, but  less than 128. Higher-valued codes never match escapes such as \ew or \ed, but
761  can be tested with \ep if PCRE is built with Unicode character property  can be tested with \ep if PCRE is built with Unicode character property
762  support.)  support. The use of locales with Unicode is discouraged.
763  .P  .P
764  An internal set of tables is created in the default C locale when PCRE is  An internal set of tables is created in the default C locale when PCRE is
765  built. This is used when the final argument of \fBpcre_compile()\fP is NULL,  built. This is used when the final argument of \fBpcre_compile()\fP is NULL,
# Line 594  check against passing an arbitrary memor Line 825  check against passing an arbitrary memor
825  \fBpcre_fullinfo()\fP, to obtain the length of the compiled pattern:  \fBpcre_fullinfo()\fP, to obtain the length of the compiled pattern:
826  .sp  .sp
827    int rc;    int rc;
828    unsigned long int length;    size_t length;
829    rc = pcre_fullinfo(    rc = pcre_fullinfo(
830      re,               /* result of pcre_compile() */      re,               /* result of pcre_compile() */
831      pe,               /* result of pcre_study(), or NULL */      pe,               /* result of pcre_study(), or NULL */
# Line 615  no back references. Line 846  no back references.
846  Return the number of capturing subpatterns in the pattern. The fourth argument  Return the number of capturing subpatterns in the pattern. The fourth argument
847  should point to an \fBint\fP variable.  should point to an \fBint\fP variable.
848  .sp  .sp
849    PCRE_INFO_DEFAULTTABLES    PCRE_INFO_DEFAULT_TABLES
850  .sp  .sp
851  Return a pointer to the internal default character tables within PCRE. The  Return a pointer to the internal default character tables within PCRE. The
852  fourth argument should point to an \fBunsigned char *\fP variable. This  fourth argument should point to an \fBunsigned char *\fP variable. This
# Line 626  a NULL table pointer. Line 857  a NULL table pointer.
857    PCRE_INFO_FIRSTBYTE    PCRE_INFO_FIRSTBYTE
858  .sp  .sp
859  Return information about the first byte of any matched string, for a  Return information about the first byte of any matched string, for a
860  non-anchored pattern. (This option used to be called PCRE_INFO_FIRSTCHAR; the  non-anchored pattern. The fourth argument should point to an \fBint\fP
861  old name is still recognized for backwards compatibility.)  variable. (This option used to be called PCRE_INFO_FIRSTCHAR; the old name is
862    still recognized for backwards compatibility.)
863  .P  .P
864  If there is a fixed first byte, for example, from a pattern such as  If there is a fixed first byte, for example, from a pattern such as
865  (cat|cow|coyote), it is returned in the integer pointed to by \fIwhere\fP.  (cat|cow|coyote), its value is returned. Otherwise, if either
 Otherwise, if either  
866  .sp  .sp
867  (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
868  starts with "^", or  starts with "^", or
# Line 666  is -1. Line 897  is -1.
897  .sp  .sp
898  PCRE supports the use of named as well as numbered capturing parentheses. The  PCRE supports the use of named as well as numbered capturing parentheses. The
899  names are just an additional way of identifying the parentheses, which still  names are just an additional way of identifying the parentheses, which still
900  acquire numbers. A convenience function called \fBpcre_get_named_substring()\fP  acquire numbers. Several convenience functions such as
901  is provided for extracting an individual captured substring by name. It is also  \fBpcre_get_named_substring()\fP are provided for extracting captured
902  possible to extract the data directly, by first converting the name to a number  substrings by name. It is also possible to extract the data directly, by first
903  in order to access the correct pointers in the output vector (described with  converting the name to a number in order to access the correct pointers in the
904  \fBpcre_exec()\fP below). To do the conversion, you need to use the  output vector (described with \fBpcre_exec()\fP below). To do the conversion,
905  name-to-number map, which is described by these three values.  you need to use the name-to-number map, which is described by these three
906    values.
907  .P  .P
908  The map consists of a number of fixed-size entries. PCRE_INFO_NAMECOUNT gives  The map consists of a number of fixed-size entries. PCRE_INFO_NAMECOUNT gives
909  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
# Line 680  length of the longest name. PCRE_INFO_NA Line 912  length of the longest name. PCRE_INFO_NA
912  entry of the table (a pointer to \fBchar\fP). The first two bytes of each entry  entry of the table (a pointer to \fBchar\fP). The first two bytes of each entry
913  are the number of the capturing parenthesis, most significant byte first. The  are the number of the capturing parenthesis, most significant byte first. The
914  rest of the entry is the corresponding name, zero terminated. The names are in  rest of the entry is the corresponding name, zero terminated. The names are in
915  alphabetical order. For example, consider the following pattern (assume  alphabetical order. When PCRE_DUPNAMES is set, duplicate names are in order of
916    their parentheses numbers. For example, consider the following pattern (assume
917  PCRE_EXTENDED is set, so white space - including newlines - is ignored):  PCRE_EXTENDED is set, so white space - including newlines - is ignored):
918  .sp  .sp
919  .\" JOIN  .\" JOIN
920    (?P<date> (?P<year>(\ed\ed)?\ed\ed) -    (?<date> (?<year>(\ed\ed)?\ed\ed) -
921    (?P<month>\ed\ed) - (?P<day>\ed\ed) )    (?<month>\ed\ed) - (?<day>\ed\ed) )
922  .sp  .sp
923  There are four named subpatterns, so the table has four entries, and each entry  There are four named subpatterns, so the table has four entries, and each entry
924  in the table is eight bytes long. The table is as follows, with non-printing  in the table is eight bytes long. The table is as follows, with non-printing
# Line 697  bytes shows in hexadecimal, and undefine Line 930  bytes shows in hexadecimal, and undefine
930    00 02 y  e  a  r  00 ??    00 02 y  e  a  r  00 ??
931  .sp  .sp
932  When writing code to extract data from named subpatterns using the  When writing code to extract data from named subpatterns using the
933  name-to-number map, remember that the length of each entry is likely to be  name-to-number map, remember that the length of the entries is likely to be
934  different for each compiled pattern.  different for each compiled pattern.
935  .sp  .sp
936    PCRE_INFO_OPTIONS    PCRE_INFO_OPTIONS
# Line 760  it is used to pass back information abou Line 993  it is used to pass back information abou
993  string (see PCRE_INFO_FIRSTBYTE above).  string (see PCRE_INFO_FIRSTBYTE above).
994  .  .
995  .  .
996  .SH "MATCHING A PATTERN"  .SH "REFERENCE COUNTS"
997    .rs
998    .sp
999    .B int pcre_refcount(pcre *\fIcode\fP, int \fIadjust\fP);
1000    .PP
1001    The \fBpcre_refcount()\fP function is used to maintain a reference count in the
1002    data block that contains a compiled pattern. It is provided for the benefit of
1003    applications that operate in an object-oriented manner, where different parts
1004    of the application may be using the same compiled pattern, but you want to free
1005    the block when they are all done.
1006    .P
1007    When a pattern is compiled, the reference count field is initialized to zero.
1008    It is changed only by calling this function, whose action is to add the
1009    \fIadjust\fP value (which may be positive or negative) to it. The yield of the
1010    function is the new value. However, the value of the count is constrained to
1011    lie between 0 and 65535, inclusive. If the new value is outside these limits,
1012    it is forced to the appropriate limit value.
1013    .P
1014    Except when it is zero, the reference count is not correctly preserved if a
1015    pattern is compiled on one host and then transferred to a host whose byte-order
1016    is different. (This seems a highly unlikely scenario.)
1017    .
1018    .
1019    .SH "MATCHING A PATTERN: THE TRADITIONAL FUNCTION"
1020  .rs  .rs
1021  .sp  .sp
1022  .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,"
# Line 772  string (see PCRE_INFO_FIRSTBYTE above). Line 1028  string (see PCRE_INFO_FIRSTBYTE above).
1028  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
1029  compiled pattern, which is passed in the \fIcode\fP argument. If the  compiled pattern, which is passed in the \fIcode\fP argument. If the
1030  pattern has been studied, the result of the study should be passed in the  pattern has been studied, the result of the study should be passed in the
1031  \fIextra\fP argument.  \fIextra\fP argument. This function is the main matching facility of the
1032    library, and it operates in a Perl-like manner. For specialist use there is
1033    also an alternative matching function, which is described
1034    .\" HTML <a href="#dfamatch">
1035    .\" </a>
1036    below
1037    .\"
1038    in the section about the \fBpcre_dfa_exec()\fP function.
1039  .P  .P
1040  In most applications, the pattern will have been compiled (and optionally  In most applications, the pattern will have been compiled (and optionally
1041  studied) in the same process that calls \fBpcre_exec()\fP. However, it is  studied) in the same process that calls \fBpcre_exec()\fP. However, it is
# Line 796  Here is an example of a simple call to \ Line 1059  Here is an example of a simple call to \
1059      0,              /* start at offset 0 in the subject */      0,              /* start at offset 0 in the subject */
1060      0,              /* default options */      0,              /* default options */
1061      ovector,        /* vector of integers for substring information */      ovector,        /* vector of integers for substring information */
1062      30);            /* number of elements in the vector (NOT size in bytes) */      30);            /* number of elements (NOT size in bytes) */
1063  .  .
1064  .\" HTML <a name="extradata"></a>  .\" HTML <a name="extradata"></a>
1065  .SS "Extra data for \fBpcre_exec()\fR"  .SS "Extra data for \fBpcre_exec()\fR"
# Line 805  Here is an example of a simple call to \ Line 1068  Here is an example of a simple call to \
1068  If the \fIextra\fP argument is not NULL, it must point to a \fBpcre_extra\fP  If the \fIextra\fP argument is not NULL, it must point to a \fBpcre_extra\fP
1069  data block. The \fBpcre_study()\fP function returns such a block (when it  data block. The \fBpcre_study()\fP function returns such a block (when it
1070  doesn't return NULL), but you can also create one for yourself, and pass  doesn't return NULL), but you can also create one for yourself, and pass
1071  additional information in it. The fields in a \fBpcre_extra\fP block are as  additional information in it. The \fBpcre_extra\fP block contains the following
1072  follows:  fields (not necessarily in this order):
1073  .sp  .sp
1074    unsigned long int \fIflags\fP;    unsigned long int \fIflags\fP;
1075    void *\fIstudy_data\fP;    void *\fIstudy_data\fP;
1076    unsigned long int \fImatch_limit\fP;    unsigned long int \fImatch_limit\fP;
1077      unsigned long int \fImatch_limit_recursion\fP;
1078    void *\fIcallout_data\fP;    void *\fIcallout_data\fP;
1079    const unsigned char *\fItables\fP;    const unsigned char *\fItables\fP;
1080  .sp  .sp
# Line 819  are set. The flag bits are: Line 1083  are set. The flag bits are:
1083  .sp  .sp
1084    PCRE_EXTRA_STUDY_DATA    PCRE_EXTRA_STUDY_DATA
1085    PCRE_EXTRA_MATCH_LIMIT    PCRE_EXTRA_MATCH_LIMIT
1086      PCRE_EXTRA_MATCH_LIMIT_RECURSION
1087    PCRE_EXTRA_CALLOUT_DATA    PCRE_EXTRA_CALLOUT_DATA
1088    PCRE_EXTRA_TABLES    PCRE_EXTRA_TABLES
1089  .sp  .sp
# Line 833  but which have a very large number of po Line 1098  but which have a very large number of po
1098  classic example is the use of nested unlimited repeats.  classic example is the use of nested unlimited repeats.
1099  .P  .P
1100  Internally, PCRE uses a function called \fBmatch()\fP which it calls repeatedly  Internally, PCRE uses a function called \fBmatch()\fP which it calls repeatedly
1101  (sometimes recursively). The limit is imposed on the number of times this  (sometimes recursively). The limit set by \fImatch_limit\fP is imposed on the
1102  function is called during a match, which has the effect of limiting the amount  number of times this function is called during a match, which has the effect of
1103  of recursion and backtracking that can take place. For patterns that are not  limiting the amount of backtracking that can take place. For patterns that are
1104  anchored, the count starts from zero for each position in the subject string.  not anchored, the count restarts from zero for each position in the subject
1105    string.
1106  .P  .P
1107  The default limit for the library can be set when PCRE is built; the default  The default value for the limit can be set when PCRE is built; the default
1108  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
1109  reduce the default by suppling \fBpcre_exec()\fP with a \fBpcre_extra\fP block  override the default by suppling \fBpcre_exec()\fP with a \fBpcre_extra\fP
1110  in which \fImatch_limit\fP is set to a smaller value, and  block in which \fImatch_limit\fP is set, and PCRE_EXTRA_MATCH_LIMIT is set in
1111  PCRE_EXTRA_MATCH_LIMIT is set in the \fIflags\fP field. If the limit is  the \fIflags\fP field. If the limit is exceeded, \fBpcre_exec()\fP returns
1112  exceeded, \fBpcre_exec()\fP returns PCRE_ERROR_MATCHLIMIT.  PCRE_ERROR_MATCHLIMIT.
1113    .P
1114    The \fImatch_limit_recursion\fP field is similar to \fImatch_limit\fP, but
1115    instead of limiting the total number of times that \fBmatch()\fP is called, it
1116    limits the depth of recursion. The recursion depth is a smaller number than the
1117    total number of calls, because not all calls to \fBmatch()\fP are recursive.
1118    This limit is of use only if it is set smaller than \fImatch_limit\fP.
1119    .P
1120    Limiting the recursion depth limits the amount of stack that can be used, or,
1121    when PCRE has been compiled to use memory on the heap instead of the stack, the
1122    amount of heap memory that can be used.
1123    .P
1124    The default value for \fImatch_limit_recursion\fP can be set when PCRE is
1125    built; the default default is the same value as the default for
1126    \fImatch_limit\fP. You can override the default by suppling \fBpcre_exec()\fP
1127    with a \fBpcre_extra\fP block in which \fImatch_limit_recursion\fP is set, and
1128    PCRE_EXTRA_MATCH_LIMIT_RECURSION is set in the \fIflags\fP field. If the limit
1129    is exceeded, \fBpcre_exec()\fP returns PCRE_ERROR_RECURSIONLIMIT.
1130  .P  .P
1131  The \fIpcre_callout\fP field is used in conjunction with the "callout" feature,  The \fIpcre_callout\fP field is used in conjunction with the "callout" feature,
1132  which is described in the  which is described in the
# Line 870  documentation for a discussion of saving Line 1153  documentation for a discussion of saving
1153  .rs  .rs
1154  .sp  .sp
1155  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
1156  zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NOTBOL,  zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_\fIxxx\fP,
1157  PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NO_UTF8_CHECK and PCRE_PARTIAL.  PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NO_UTF8_CHECK and PCRE_PARTIAL.
1158  .sp  .sp
1159    PCRE_ANCHORED    PCRE_ANCHORED
1160  .sp  .sp
# Line 880  matching position. If a pattern was comp Line 1163  matching position. If a pattern was comp
1163  to be anchored by virtue of its contents, it cannot be made unachored at  to be anchored by virtue of its contents, it cannot be made unachored at
1164  matching time.  matching time.
1165  .sp  .sp
1166      PCRE_NEWLINE_CR
1167      PCRE_NEWLINE_LF
1168      PCRE_NEWLINE_CRLF
1169      PCRE_NEWLINE_ANY
1170    .sp
1171    These options override the newline definition that was chosen or defaulted when
1172    the pattern was compiled. For details, see the description of
1173    \fBpcre_compile()\fP above. During matching, the newline choice affects the
1174    behaviour of the dot, circumflex, and dollar metacharacters. It may also alter
1175    the way the match position is advanced after a match failure for an unanchored
1176    pattern. When PCRE_NEWLINE_CRLF or PCRE_NEWLINE_ANY is set, and a match attempt
1177    fails when the current position is at a CRLF sequence, the match position is
1178    advanced by two characters instead of one, in other words, to after the CRLF.
1179    .sp
1180    PCRE_NOTBOL    PCRE_NOTBOL
1181  .sp  .sp
1182  This option specifies that first character of the subject string is not the  This option specifies that first character of the subject string is not the
# Line 1011  is set to the offset of the first charac Line 1308  is set to the offset of the first charac
1308  first pair, \fIovector[0]\fP and \fIovector[1]\fP, identify the portion of the  first pair, \fIovector[0]\fP and \fIovector[1]\fP, identify the portion of the
1309  subject string matched by the entire pattern. The next pair is used for the  subject string matched by the entire pattern. The next pair is used for the
1310  first capturing subpattern, and so on. The value returned by \fBpcre_exec()\fP  first capturing subpattern, and so on. The value returned by \fBpcre_exec()\fP
1311  is the number of pairs that have been set. If there are no capturing  is one more than the highest numbered pair that has been set. For example, if
1312  subpatterns, the return value from a successful match is 1, indicating that  two substrings have been captured, the returned value is 3. If there are no
1313  just the first pair of offsets has been set.  capturing subpatterns, the return value from a successful match is 1,
1314  .P  indicating that just the first pair of offsets has been set.
 Some convenience functions are provided for extracting the captured substrings  
 as separate strings. These are described in the following section.  
 .P  
 It is possible for an capturing subpattern number \fIn+1\fP to match some  
 part of the subject when subpattern \fIn\fP has not been used at all. For  
 example, if the string "abc" is matched against the pattern (a|(z))(bc)  
 subpatterns 1 and 3 are matched, but 2 is not. When this happens, both offset  
 values corresponding to the unused subpattern are set to -1.  
1315  .P  .P
1316  If a capturing subpattern is matched repeatedly, it is the last portion of the  If a capturing subpattern is matched repeatedly, it is the last portion of the
1317  string that it matched that is returned.  string that it matched that is returned.
# Line 1036  the \fIovector\fP is not big enough to r Line 1325  the \fIovector\fP is not big enough to r
1325  has to get additional memory for use during matching. Thus it is usually  has to get additional memory for use during matching. Thus it is usually
1326  advisable to supply an \fIovector\fP.  advisable to supply an \fIovector\fP.
1327  .P  .P
1328  Note that \fBpcre_info()\fP can be used to find out how many capturing  The \fBpcre_info()\fP function can be used to find out how many capturing
1329  subpatterns there are in a compiled pattern. The smallest size for  subpatterns there are in a compiled pattern. The smallest size for
1330  \fIovector\fP that will allow for \fIn\fP captured substrings, in addition to  \fIovector\fP that will allow for \fIn\fP captured substrings, in addition to
1331  the offsets of the substring matched by the whole pattern, is (\fIn\fP+1)*3.  the offsets of the substring matched by the whole pattern, is (\fIn\fP+1)*3.
1332    .P
1333    It is possible for capturing subpattern number \fIn+1\fP to match some part of
1334    the subject when subpattern \fIn\fP has not been used at all. For example, if
1335    the string "abc" is matched against the pattern (a|(z))(bc) the return from the
1336    function is 4, and subpatterns 1 and 3 are matched, but 2 is not. When this
1337    happens, both values in the offset pairs corresponding to unused subpatterns
1338    are set to -1.
1339    .P
1340    Offset values that correspond to unused subpatterns at the end of the
1341    expression are also set to -1. For example, if the string "abc" is matched
1342    against the pattern (abc)(x(yz)?)? subpatterns 2 and 3 are not matched. The
1343    return from the function is 2, because the highest used capturing subpattern
1344    number is 1. However, you can refer to the offsets for the second and third
1345    capturing subpatterns if you wish (assuming the vector is large enough, of
1346    course).
1347    .P
1348    Some convenience functions are provided for extracting the captured substrings
1349    as separate strings. These are described below.
1350  .  .
1351  .SS "Return values from \fBpcre_exec()\fP"  .\" HTML <a name="errorlist"></a>
1352    .SS "Error return values from \fBpcre_exec()\fP"
1353  .rs  .rs
1354  .sp  .sp
1355  If \fBpcre_exec()\fP fails, it returns a negative number. The following are  If \fBpcre_exec()\fP fails, it returns a negative number. The following are
# Line 1068  compiled in an environment of one endian Line 1376  compiled in an environment of one endian
1376  other endianness. This is the error that PCRE gives when the magic number is  other endianness. This is the error that PCRE gives when the magic number is
1377  not present.  not present.
1378  .sp  .sp
1379    PCRE_ERROR_UNKNOWN_NODE   (-5)    PCRE_ERROR_UNKNOWN_OPCODE (-5)
1380  .sp  .sp
1381  While running the pattern match, an unknown item was encountered in the  While running the pattern match, an unknown item was encountered in the
1382  compiled pattern. This error could be caused by a bug in PCRE or by overwriting  compiled pattern. This error could be caused by a bug in PCRE or by overwriting
# Line 1090  below). It is never returned by \fBpcre_ Line 1398  below). It is never returned by \fBpcre_
1398  .sp  .sp
1399    PCRE_ERROR_MATCHLIMIT     (-8)    PCRE_ERROR_MATCHLIMIT     (-8)
1400  .sp  .sp
1401  The recursion and backtracking limit, as specified by the \fImatch_limit\fP  The backtracking limit, as specified by the \fImatch_limit\fP field in a
1402  field in a \fBpcre_extra\fP structure (or defaulted) was reached. See the  \fBpcre_extra\fP structure (or defaulted) was reached. See the description
1403  description above.  above.
1404  .sp  .sp
1405    PCRE_ERROR_CALLOUT        (-9)    PCRE_ERROR_CALLOUT        (-9)
1406  .sp  .sp
# Line 1112  A string that contains an invalid UTF-8 Line 1420  A string that contains an invalid UTF-8
1420  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 valid, but the value
1421  of \fIstartoffset\fP did not point to the beginning of a UTF-8 character.  of \fIstartoffset\fP did not point to the beginning of a UTF-8 character.
1422  .sp  .sp
1423    PCRE_ERROR_PARTIAL (-12)    PCRE_ERROR_PARTIAL        (-12)
1424  .sp  .sp
1425  The subject string did not match, but it did match partially. See the  The subject string did not match, but it did match partially. See the
1426  .\" HREF  .\" HREF
# Line 1120  The subject string did not match, but it Line 1428  The subject string did not match, but it
1428  .\"  .\"
1429  documentation for details of partial matching.  documentation for details of partial matching.
1430  .sp  .sp
1431    PCRE_ERROR_BAD_PARTIAL (-13)    PCRE_ERROR_BADPARTIAL     (-13)
1432  .sp  .sp
1433  The PCRE_PARTIAL option was used with a compiled pattern containing items that  The PCRE_PARTIAL option was used with a compiled pattern containing items that
1434  are not supported for partial matching. See the  are not supported for partial matching. See the
# Line 1129  are not supported for partial matching. Line 1437  are not supported for partial matching.
1437  .\"  .\"
1438  documentation for details of partial matching.  documentation for details of partial matching.
1439  .sp  .sp
1440    PCRE_ERROR_INTERNAL (-14)    PCRE_ERROR_INTERNAL       (-14)
1441  .sp  .sp
1442  An unexpected internal error has occurred. This error could be caused by a bug  An unexpected internal error has occurred. This error could be caused by a bug
1443  in PCRE or by overwriting of the compiled pattern.  in PCRE or by overwriting of the compiled pattern.
1444  .sp  .sp
1445    PCRE_ERROR_BADCOUNT (-15)    PCRE_ERROR_BADCOUNT       (-15)
1446  .sp  .sp
1447  This error is given if the value of the \fIovecsize\fP argument is negative.  This error is given if the value of the \fIovecsize\fP argument is negative.
1448    .sp
1449      PCRE_ERROR_RECURSIONLIMIT (-21)
1450    .sp
1451    The internal recursion limit, as specified by the \fImatch_limit_recursion\fP
1452    field in a \fBpcre_extra\fP structure (or defaulted) was reached. See the
1453    description above.
1454    .sp
1455      PCRE_ERROR_NULLWSLIMIT    (-22)
1456    .sp
1457    When a group that can match an empty substring is repeated with an unbounded
1458    upper limit, the subject position at the start of the group must be remembered,
1459    so that a test for an empty string can be made when the end of the group is
1460    reached. Some workspace is required for this; if it runs out, this error is
1461    given.
1462    .sp
1463      PCRE_ERROR_BADNEWLINE     (-23)
1464    .sp
1465    An invalid combination of PCRE_NEWLINE_\fIxxx\fP options was given.
1466    .P
1467    Error numbers -16 to -20 are not used by \fBpcre_exec()\fP.
1468  .  .
1469  .  .
1470  .SH "EXTRACTING CAPTURED SUBSTRINGS BY NUMBER"  .SH "EXTRACTING CAPTURED SUBSTRINGS BY NUMBER"
# Line 1166  Captured substrings can be accessed dire Line 1494  Captured substrings can be accessed dire
1494  \fBpcre_get_substring_list()\fP are provided for extracting captured substrings  \fBpcre_get_substring_list()\fP are provided for extracting captured substrings
1495  as new, separate, zero-terminated strings. These functions identify substrings  as new, separate, zero-terminated strings. These functions identify substrings
1496  by number. The next section describes functions for extracting named  by number. The next section describes functions for extracting named
1497  substrings. A substring that contains a binary zero is correctly extracted and  substrings.
1498  has a further zero added on the end, but the result is not, of course,  .P
1499  a C string.  A substring that contains a binary zero is correctly extracted and has a
1500    further zero added on the end, but the result is not, of course, a C string.
1501    However, you can process such a string by referring to the length that is
1502    returned by \fBpcre_copy_substring()\fP and \fBpcre_get_substring()\fP.
1503    Unfortunately, the interface to \fBpcre_get_substring_list()\fP is not adequate
1504    for handling strings containing binary zeros, because the end of the final
1505    string is not independently indicated.
1506  .P  .P
1507  The first three arguments are the same for all three of these functions:  The first three arguments are the same for all three of these functions:
1508  \fIsubject\fP is the subject string that has just been successfully matched,  \fIsubject\fP is the subject string that has just been successfully matched,
# Line 1188  the string is placed in \fIbuffer\fP, wh Line 1522  the string is placed in \fIbuffer\fP, wh
1522  \fIbuffersize\fP, while for \fBpcre_get_substring()\fP a new block of memory is  \fIbuffersize\fP, while for \fBpcre_get_substring()\fP a new block of memory is
1523  obtained via \fBpcre_malloc\fP, and its address is returned via  obtained via \fBpcre_malloc\fP, and its address is returned via
1524  \fIstringptr\fP. The yield of the function is the length of the string, not  \fIstringptr\fP. The yield of the function is the length of the string, not
1525  including the terminating zero, or one of  including the terminating zero, or one of these error codes:
1526  .sp  .sp
1527    PCRE_ERROR_NOMEMORY       (-6)    PCRE_ERROR_NOMEMORY       (-6)
1528  .sp  .sp
# Line 1204  and builds a list of pointers to them. A Line 1538  and builds a list of pointers to them. A
1538  memory that is obtained via \fBpcre_malloc\fP. The address of the memory block  memory that is obtained via \fBpcre_malloc\fP. The address of the memory block
1539  is returned via \fIlistptr\fP, which is also the start of the list of string  is returned via \fIlistptr\fP, which is also the start of the list of string
1540  pointers. The end of the list is marked by a NULL pointer. The yield of the  pointers. The end of the list is marked by a NULL pointer. The yield of the
1541  function is zero if all went well, or  function is zero if all went well, or the error code
1542  .sp  .sp
1543    PCRE_ERROR_NOMEMORY       (-6)    PCRE_ERROR_NOMEMORY       (-6)
1544  .sp  .sp
# Line 1223  a previous call of \fBpcre_get_substring Line 1557  a previous call of \fBpcre_get_substring
1557  \fBpcre_get_substring_list()\fP, respectively. They do nothing more than call  \fBpcre_get_substring_list()\fP, respectively. They do nothing more than call
1558  the function pointed to by \fBpcre_free\fP, which of course could be called  the function pointed to by \fBpcre_free\fP, which of course could be called
1559  directly from a C program. However, PCRE is used in some situations where it is  directly from a C program. However, PCRE is used in some situations where it is
1560  linked via a special interface to another programming language which cannot use  linked via a special interface to another programming language that cannot use
1561  \fBpcre_free\fP directly; it is for these cases that the functions are  \fBpcre_free\fP directly; it is for these cases that the functions are
1562  provided.  provided.
1563  .  .
# Line 1258  For example, for this pattern Line 1592  For example, for this pattern
1592  .sp  .sp
1593    (a+)b(?<xxx>\ed+)...    (a+)b(?<xxx>\ed+)...
1594  .sp  .sp
1595  the number of the subpattern called "xxx" is 2. You can find the number from  the number of the subpattern called "xxx" is 2. If the name is known to be
1596  the name by calling \fBpcre_get_stringnumber()\fP. The first argument is the  unique (PCRE_DUPNAMES was not set), you can find the number from the name by
1597  compiled pattern, and the second is the name. The yield of the function is the  calling \fBpcre_get_stringnumber()\fP. The first argument is the compiled
1598    pattern, and the second is the name. The yield of the function is the
1599  subpattern number, or PCRE_ERROR_NOSUBSTRING (-7) if there is no subpattern of  subpattern number, or PCRE_ERROR_NOSUBSTRING (-7) if there is no subpattern of
1600  that name.  that name.
1601  .P  .P
# Line 1268  Given the number, you can extract the su Line 1603  Given the number, you can extract the su
1603  functions described in the previous section. For convenience, there are also  functions described in the previous section. For convenience, there are also
1604  two functions that do the whole job.  two functions that do the whole job.
1605  .P  .P
1606  Most of the arguments of \fIpcre_copy_named_substring()\fP and  Most of the arguments of \fBpcre_copy_named_substring()\fP and
1607  \fIpcre_get_named_substring()\fP are the same as those for the similarly named  \fBpcre_get_named_substring()\fP are the same as those for the similarly named
1608  functions that extract by number. As these are described in the previous  functions that extract by number. As these are described in the previous
1609  section, they are not re-described here. There are just two differences:  section, they are not re-described here. There are just two differences:
1610  .P  .P
# Line 1281  translation table. Line 1616  translation table.
1616  These functions call \fBpcre_get_stringnumber()\fP, and if it succeeds, they  These functions call \fBpcre_get_stringnumber()\fP, and if it succeeds, they
1617  then call \fIpcre_copy_substring()\fP or \fIpcre_get_substring()\fP, as  then call \fIpcre_copy_substring()\fP or \fIpcre_get_substring()\fP, as
1618  appropriate.  appropriate.
1619    .
1620    .
1621    .SH "DUPLICATE SUBPATTERN NAMES"
1622    .rs
1623    .sp
1624    .B int pcre_get_stringtable_entries(const pcre *\fIcode\fP,
1625    .ti +5n
1626    .B const char *\fIname\fP, char **\fIfirst\fP, char **\fIlast\fP);
1627    .PP
1628    When a pattern is compiled with the PCRE_DUPNAMES option, names for subpatterns
1629    are not required to be unique. Normally, patterns with duplicate names are such
1630    that in any one match, only one of the named subpatterns participates. An
1631    example is shown in the
1632    .\" HREF
1633    \fBpcrepattern\fP
1634    .\"
1635    documentation. When duplicates are present, \fBpcre_copy_named_substring()\fP
1636    and \fBpcre_get_named_substring()\fP return the first substring corresponding
1637    to the given name that is set. If none are set, an empty string is returned.
1638    The \fBpcre_get_stringnumber()\fP function returns one of the numbers that are
1639    associated with the name, but it is not defined which it is.
1640    .sp
1641    If you want to get full details of all captured substrings for a given name,
1642    you must use the \fBpcre_get_stringtable_entries()\fP function. The first
1643    argument is the compiled pattern, and the second is the name. The third and
1644    fourth are pointers to variables which are updated by the function. After it
1645    has run, they point to the first and last entries in the name-to-number table
1646    for the given name. The function itself returns the length of each entry, or
1647    PCRE_ERROR_NOSUBSTRING (-7) if there are none. The format of the table is
1648    described above in the section entitled \fIInformation about a pattern\fP.
1649    Given all the relevant entries for the name, you can extract each of their
1650    numbers, and hence the captured data, if any.
1651    .
1652    .
1653    .SH "FINDING ALL POSSIBLE MATCHES"
1654    .rs
1655    .sp
1656    The traditional matching function uses a similar algorithm to Perl, which stops
1657    when it finds the first match, starting at a given point in the subject. If you
1658    want to find all possible matches, or the longest possible match, consider
1659    using the alternative matching function (see below) instead. If you cannot use
1660    the alternative function, but still need to find all possible matches, you
1661    can kludge it up by making use of the callout facility, which is described in
1662    the
1663    .\" HREF
1664    \fBpcrecallout\fP
1665    .\"
1666    documentation.
1667    .P
1668    What you have to do is to insert a callout right at the end of the pattern.
1669    When your callout function is called, extract and save the current matched
1670    substring. Then return 1, which forces \fBpcre_exec()\fP to backtrack and try
1671    other alternatives. Ultimately, when it runs out of matches, \fBpcre_exec()\fP
1672    will yield PCRE_ERROR_NOMATCH.
1673    .
1674    .
1675    .\" HTML <a name="dfamatch"></a>
1676    .SH "MATCHING A PATTERN: THE ALTERNATIVE FUNCTION"
1677    .rs
1678    .sp
1679    .B int pcre_dfa_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
1680    .ti +5n
1681    .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
1682    .ti +5n
1683    .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP,
1684    .ti +5n
1685    .B int *\fIworkspace\fP, int \fIwscount\fP);
1686    .P
1687    The function \fBpcre_dfa_exec()\fP is called to match a subject string against
1688    a compiled pattern, using a matching algorithm that scans the subject string
1689    just once, and does not backtrack. This has different characteristics to the
1690    normal algorithm, and is not compatible with Perl. Some of the features of PCRE
1691    patterns are not supported. Nevertheless, there are times when this kind of
1692    matching can be useful. For a discussion of the two matching algorithms, see
1693    the
1694    .\" HREF
1695    \fBpcrematching\fP
1696    .\"
1697    documentation.
1698    .P
1699    The arguments for the \fBpcre_dfa_exec()\fP function are the same as for
1700    \fBpcre_exec()\fP, plus two extras. The \fIovector\fP argument is used in a
1701    different way, and this is described below. The other common arguments are used
1702    in the same way as for \fBpcre_exec()\fP, so their description is not repeated
1703    here.
1704    .P
1705    The two additional arguments provide workspace for the function. The workspace
1706    vector should contain at least 20 elements. It is used for keeping track of
1707    multiple paths through the pattern tree. More workspace will be needed for
1708    patterns and subjects where there are a lot of potential matches.
1709    .P
1710    Here is an example of a simple call to \fBpcre_dfa_exec()\fP:
1711    .sp
1712      int rc;
1713      int ovector[10];
1714      int wspace[20];
1715      rc = pcre_dfa_exec(
1716        re,             /* result of pcre_compile() */
1717        NULL,           /* we didn't study the pattern */
1718        "some string",  /* the subject string */
1719        11,             /* the length of the subject string */
1720        0,              /* start at offset 0 in the subject */
1721        0,              /* default options */
1722        ovector,        /* vector of integers for substring information */
1723        10,             /* number of elements (NOT size in bytes) */
1724        wspace,         /* working space vector */
1725        20);            /* number of elements (NOT size in bytes) */
1726    .
1727    .SS "Option bits for \fBpcre_dfa_exec()\fP"
1728    .rs
1729    .sp
1730    The unused bits of the \fIoptions\fP argument for \fBpcre_dfa_exec()\fP must be
1731    zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_\fIxxx\fP,
1732    PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NO_UTF8_CHECK, PCRE_PARTIAL,
1733    PCRE_DFA_SHORTEST, and PCRE_DFA_RESTART. All but the last three of these are
1734    the same as for \fBpcre_exec()\fP, so their description is not repeated here.
1735    .sp
1736      PCRE_PARTIAL
1737    .sp
1738    This has the same general effect as it does for \fBpcre_exec()\fP, but the
1739    details are slightly different. When PCRE_PARTIAL is set for
1740    \fBpcre_dfa_exec()\fP, the return code PCRE_ERROR_NOMATCH is converted into
1741    PCRE_ERROR_PARTIAL if the end of the subject is reached, there have been no
1742    complete matches, but there is still at least one matching possibility. The
1743    portion of the string that provided the partial match is set as the first
1744    matching string.
1745    .sp
1746      PCRE_DFA_SHORTEST
1747    .sp
1748    Setting the PCRE_DFA_SHORTEST option causes the matching algorithm to stop as
1749    soon as it has found one match. Because of the way the alternative algorithm
1750    works, this is necessarily the shortest possible match at the first possible
1751    matching point in the subject string.
1752    .sp
1753      PCRE_DFA_RESTART
1754    .sp
1755    When \fBpcre_dfa_exec()\fP is called with the PCRE_PARTIAL option, and returns
1756    a partial match, it is possible to call it again, with additional subject
1757    characters, and have it continue with the same match. The PCRE_DFA_RESTART
1758    option requests this action; when it is set, the \fIworkspace\fP and
1759    \fIwscount\fP options must reference the same vector as before because data
1760    about the match so far is left in them after a partial match. There is more
1761    discussion of this facility in the
1762    .\" HREF
1763    \fBpcrepartial\fP
1764    .\"
1765    documentation.
1766    .
1767    .SS "Successful returns from \fBpcre_dfa_exec()\fP"
1768    .rs
1769    .sp
1770    When \fBpcre_dfa_exec()\fP succeeds, it may have matched more than one
1771    substring in the subject. Note, however, that all the matches from one run of
1772    the function start at the same point in the subject. The shorter matches are
1773    all initial substrings of the longer matches. For example, if the pattern
1774    .sp
1775      <.*>
1776    .sp
1777    is matched against the string
1778    .sp
1779      This is <something> <something else> <something further> no more
1780    .sp
1781    the three matched strings are
1782    .sp
1783      <something>
1784      <something> <something else>
1785      <something> <something else> <something further>
1786    .sp
1787    On success, the yield of the function is a number greater than zero, which is
1788    the number of matched substrings. The substrings themselves are returned in
1789    \fIovector\fP. Each string uses two elements; the first is the offset to the
1790    start, and the second is the offset to the end. In fact, all the strings have
1791    the same start offset. (Space could have been saved by giving this only once,
1792    but it was decided to retain some compatibility with the way \fBpcre_exec()\fP
1793    returns data, even though the meaning of the strings is different.)
1794    .P
1795    The strings are returned in reverse order of length; that is, the longest
1796    matching string is given first. If there were too many matches to fit into
1797    \fIovector\fP, the yield of the function is zero, and the vector is filled with
1798    the longest matches.
1799    .
1800    .SS "Error returns from \fBpcre_dfa_exec()\fP"
1801    .rs
1802    .sp
1803    The \fBpcre_dfa_exec()\fP function returns a negative number when it fails.
1804    Many of the errors are the same as for \fBpcre_exec()\fP, and these are
1805    described
1806    .\" HTML <a href="#errorlist">
1807    .\" </a>
1808    above.
1809    .\"
1810    There are in addition the following errors that are specific to
1811    \fBpcre_dfa_exec()\fP:
1812    .sp
1813      PCRE_ERROR_DFA_UITEM      (-16)
1814    .sp
1815    This return is given if \fBpcre_dfa_exec()\fP encounters an item in the pattern
1816    that it does not support, for instance, the use of \eC or a back reference.
1817    .sp
1818      PCRE_ERROR_DFA_UCOND      (-17)
1819    .sp
1820    This return is given if \fBpcre_dfa_exec()\fP encounters a condition item that
1821    uses a back reference for the condition, or a test for recursion in a specific
1822    group. These are not supported.
1823    .sp
1824      PCRE_ERROR_DFA_UMLIMIT    (-18)
1825    .sp
1826    This return is given if \fBpcre_dfa_exec()\fP is called with an \fIextra\fP
1827    block that contains a setting of the \fImatch_limit\fP field. This is not
1828    supported (it is meaningless).
1829    .sp
1830      PCRE_ERROR_DFA_WSSIZE     (-19)
1831    .sp
1832    This return is given if \fBpcre_dfa_exec()\fP runs out of space in the
1833    \fIworkspace\fP vector.
1834    .sp
1835      PCRE_ERROR_DFA_RECURSE    (-20)
1836    .sp
1837    When a recursive subpattern is processed, the matching function calls itself
1838    recursively, using private vectors for \fIovector\fP and \fIworkspace\fP. This
1839    error is given if the output vector is not large enough. This should be
1840    extremely rare, as a vector of size 1000 is used.
1841    .
1842    .
1843    .SH "SEE ALSO"
1844    .rs
1845    .sp
1846    \fBpcrebuild\fP(3), \fBpcrecallout\fP(3), \fBpcrecpp(3)\fP(3),
1847    \fBpcrematching\fP(3), \fBpcrepartial\fP(3), \fBpcreposix\fP(3),
1848    \fBpcreprecompile\fP(3), \fBpcresample\fP(3), \fBpcrestack\fP(3).
1849  .P  .P
1850  .in 0  .in 0
1851  Last updated: 09 September 2004  Last updated: 30 November 2006
1852  .br  .br
1853  Copyright (c) 1997-2004 University of Cambridge.  Copyright (c) 1997-2006 University of Cambridge.

Legend:
Removed from v.75  
changed lines
  Added in v.93

  ViewVC Help
Powered by ViewVC 1.1.5