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

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

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

revision 65 by nigel, Sat Feb 24 21:40:08 2007 UTC revision 172 by ph10, Tue Jun 5 10:40:13 2007 UTC
# Line 3  Line 3 
3  <title>pcreapi specification</title>  <title>pcreapi specification</title>
4  </head>  </head>
5  <body bgcolor="#FFFFFF" text="#00005A" link="#0066FF" alink="#3399FF" vlink="#2222BB">  <body bgcolor="#FFFFFF" text="#00005A" link="#0066FF" alink="#3399FF" vlink="#2222BB">
6  This HTML document has been generated automatically from the original man page.  <h1>pcreapi man page</h1>
7  If there is any nonsense in it, please consult the man page, in case the  <p>
8  conversion went wrong.<br>  Return to the <a href="index.html">PCRE index page</a>.
9    </p>
10    <p>
11    This page is part of the PCRE HTML documentation. It was generated automatically
12    from the original man page. If there is any nonsense in it, please consult the
13    man page, in case the conversion went wrong.
14    <br>
15  <ul>  <ul>
16  <li><a name="TOC1" href="#SEC1">SYNOPSIS OF PCRE API</a>  <li><a name="TOC1" href="#SEC1">PCRE NATIVE API</a>
17  <li><a name="TOC2" href="#SEC2">PCRE API</a>  <li><a name="TOC2" href="#SEC2">PCRE API OVERVIEW</a>
18  <li><a name="TOC3" href="#SEC3">MULTITHREADING</a>  <li><a name="TOC3" href="#SEC3">NEWLINES</a>
19  <li><a name="TOC4" href="#SEC4">CHECKING BUILD-TIME OPTIONS</a>  <li><a name="TOC4" href="#SEC4">MULTITHREADING</a>
20  <li><a name="TOC5" href="#SEC5">COMPILING A PATTERN</a>  <li><a name="TOC5" href="#SEC5">SAVING PRECOMPILED PATTERNS FOR LATER USE</a>
21  <li><a name="TOC6" href="#SEC6">STUDYING A PATTERN</a>  <li><a name="TOC6" href="#SEC6">CHECKING BUILD-TIME OPTIONS</a>
22  <li><a name="TOC7" href="#SEC7">LOCALE SUPPORT</a>  <li><a name="TOC7" href="#SEC7">COMPILING A PATTERN</a>
23  <li><a name="TOC8" href="#SEC8">INFORMATION ABOUT A PATTERN</a>  <li><a name="TOC8" href="#SEC8">COMPILATION ERROR CODES</a>
24  <li><a name="TOC9" href="#SEC9">OBSOLETE INFO FUNCTION</a>  <li><a name="TOC9" href="#SEC9">STUDYING A PATTERN</a>
25  <li><a name="TOC10" href="#SEC10">MATCHING A PATTERN</a>  <li><a name="TOC10" href="#SEC10">LOCALE SUPPORT</a>
26  <li><a name="TOC11" href="#SEC11">EXTRACTING CAPTURED SUBSTRINGS BY NUMBER</a>  <li><a name="TOC11" href="#SEC11">INFORMATION ABOUT A PATTERN</a>
27  <li><a name="TOC12" href="#SEC12">EXTRACTING CAPTURED SUBSTRINGS BY NAME</a>  <li><a name="TOC12" href="#SEC12">OBSOLETE INFO FUNCTION</a>
28    <li><a name="TOC13" href="#SEC13">REFERENCE COUNTS</a>
29    <li><a name="TOC14" href="#SEC14">MATCHING A PATTERN: THE TRADITIONAL FUNCTION</a>
30    <li><a name="TOC15" href="#SEC15">EXTRACTING CAPTURED SUBSTRINGS BY NUMBER</a>
31    <li><a name="TOC16" href="#SEC16">EXTRACTING CAPTURED SUBSTRINGS BY NAME</a>
32    <li><a name="TOC17" href="#SEC17">DUPLICATE SUBPATTERN NAMES</a>
33    <li><a name="TOC18" href="#SEC18">FINDING ALL POSSIBLE MATCHES</a>
34    <li><a name="TOC19" href="#SEC19">MATCHING A PATTERN: THE ALTERNATIVE FUNCTION</a>
35    <li><a name="TOC20" href="#SEC20">SEE ALSO</a>
36    <li><a name="TOC21" href="#SEC21">AUTHOR</a>
37    <li><a name="TOC22" href="#SEC22">REVISION</a>
38  </ul>  </ul>
39  <br><a name="SEC1" href="#TOC1">SYNOPSIS OF PCRE API</a><br>  <br><a name="SEC1" href="#TOC1">PCRE NATIVE API</a><br>
40  <P>  <P>
41  <b>#include &#60;pcre.h&#62;</b>  <b>#include &#60;pcre.h&#62;</b>
42  </P>  </P>
# Line 30  conversion went wrong.<br> Line 46  conversion went wrong.<br>
46  <b>const unsigned char *<i>tableptr</i>);</b>  <b>const unsigned char *<i>tableptr</i>);</b>
47  </P>  </P>
48  <P>  <P>
49    <b>pcre *pcre_compile2(const char *<i>pattern</i>, int <i>options</i>,</b>
50    <b>int *<i>errorcodeptr</i>,</b>
51    <b>const char **<i>errptr</i>, int *<i>erroffset</i>,</b>
52    <b>const unsigned char *<i>tableptr</i>);</b>
53    </P>
54    <P>
55  <b>pcre_extra *pcre_study(const pcre *<i>code</i>, int <i>options</i>,</b>  <b>pcre_extra *pcre_study(const pcre *<i>code</i>, int <i>options</i>,</b>
56  <b>const char **<i>errptr</i>);</b>  <b>const char **<i>errptr</i>);</b>
57  </P>  </P>
# Line 39  conversion went wrong.<br> Line 61  conversion went wrong.<br>
61  <b>int <i>options</i>, int *<i>ovector</i>, int <i>ovecsize</i>);</b>  <b>int <i>options</i>, int *<i>ovector</i>, int <i>ovecsize</i>);</b>
62  </P>  </P>
63  <P>  <P>
64    <b>int pcre_dfa_exec(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>
65    <b>const char *<i>subject</i>, int <i>length</i>, int <i>startoffset</i>,</b>
66    <b>int <i>options</i>, int *<i>ovector</i>, int <i>ovecsize</i>,</b>
67    <b>int *<i>workspace</i>, int <i>wscount</i>);</b>
68    </P>
69    <P>
70  <b>int pcre_copy_named_substring(const pcre *<i>code</i>,</b>  <b>int pcre_copy_named_substring(const pcre *<i>code</i>,</b>
71  <b>const char *<i>subject</i>, int *<i>ovector</i>,</b>  <b>const char *<i>subject</i>, int *<i>ovector</i>,</b>
72  <b>int <i>stringcount</i>, const char *<i>stringname</i>,</b>  <b>int <i>stringcount</i>, const char *<i>stringname</i>,</b>
# Line 60  conversion went wrong.<br> Line 88  conversion went wrong.<br>
88  <b>const char *<i>name</i>);</b>  <b>const char *<i>name</i>);</b>
89  </P>  </P>
90  <P>  <P>
91    <b>int pcre_get_stringtable_entries(const pcre *<i>code</i>,</b>
92    <b>const char *<i>name</i>, char **<i>first</i>, char **<i>last</i>);</b>
93    </P>
94    <P>
95  <b>int pcre_get_substring(const char *<i>subject</i>, int *<i>ovector</i>,</b>  <b>int pcre_get_substring(const char *<i>subject</i>, int *<i>ovector</i>,</b>
96  <b>int <i>stringcount</i>, int <i>stringnumber</i>,</b>  <b>int <i>stringcount</i>, int <i>stringnumber</i>,</b>
97  <b>const char **<i>stringptr</i>);</b>  <b>const char **<i>stringptr</i>);</b>
# Line 86  conversion went wrong.<br> Line 118  conversion went wrong.<br>
118  <b>*<i>firstcharptr</i>);</b>  <b>*<i>firstcharptr</i>);</b>
119  </P>  </P>
120  <P>  <P>
121    <b>int pcre_refcount(pcre *<i>code</i>, int <i>adjust</i>);</b>
122    </P>
123    <P>
124  <b>int pcre_config(int <i>what</i>, void *<i>where</i>);</b>  <b>int pcre_config(int <i>what</i>, void *<i>where</i>);</b>
125  </P>  </P>
126  <P>  <P>
# Line 98  conversion went wrong.<br> Line 133  conversion went wrong.<br>
133  <b>void (*pcre_free)(void *);</b>  <b>void (*pcre_free)(void *);</b>
134  </P>  </P>
135  <P>  <P>
136  <b>int (*pcre_callout)(pcre_callout_block *);</b>  <b>void *(*pcre_stack_malloc)(size_t);</b>
137  </P>  </P>
 <br><a name="SEC2" href="#TOC1">PCRE API</a><br>  
138  <P>  <P>
139  PCRE has its own native API, which is described in this document. There is also  <b>void (*pcre_stack_free)(void *);</b>
 a set of wrapper functions that correspond to the POSIX regular expression API.  
 These are described in the <b>pcreposix</b> documentation.  
140  </P>  </P>
141  <P>  <P>
142  The native API function prototypes are defined in the header file <b>pcre.h</b>,  <b>int (*pcre_callout)(pcre_callout_block *);</b>
 and on Unix systems the library itself is called <b>libpcre.a</b>, so can be  
 accessed by adding <b>-lpcre</b> to the command for linking an application which  
 calls it. The header file defines the macros PCRE_MAJOR and PCRE_MINOR to  
 contain the major and minor release numbers for the library. Applications can  
 use these to include support for different releases.  
143  </P>  </P>
144    <br><a name="SEC2" href="#TOC1">PCRE API OVERVIEW</a><br>
145  <P>  <P>
146  The functions <b>pcre_compile()</b>, <b>pcre_study()</b>, and <b>pcre_exec()</b>  PCRE has its own native API, which is described in this document. There are
147  are used for compiling and matching regular expressions. A sample program that  also some wrapper functions that correspond to the POSIX regular expression
148  demonstrates the simplest way of using them is given in the file  API. These are described in the
149  <i>pcredemo.c</i>. The <b>pcresample</b> documentation describes how to run it.  <a href="pcreposix.html"><b>pcreposix</b></a>
150    documentation. Both of these APIs define a set of C function calls. A C++
151    wrapper is distributed with PCRE. It is documented in the
152    <a href="pcrecpp.html"><b>pcrecpp</b></a>
153    page.
154  </P>  </P>
155  <P>  <P>
156  There are convenience functions for extracting captured substrings from a  The native API C function prototypes are defined in the header file
157  matched subject string. They are:  <b>pcre.h</b>, and on Unix systems the library itself is called <b>libpcre</b>.
158    It can normally be accessed by adding <b>-lpcre</b> to the command for linking
159    an application that uses PCRE. The header file defines the macros PCRE_MAJOR
160    and PCRE_MINOR to contain the major and minor release numbers for the library.
161    Applications can use these to include support for different releases of PCRE.
162    </P>
163    <P>
164    The functions <b>pcre_compile()</b>, <b>pcre_compile2()</b>, <b>pcre_study()</b>,
165    and <b>pcre_exec()</b> are used for compiling and matching regular expressions
166    in a Perl-compatible manner. A sample program that demonstrates the simplest
167    way of using them is provided in the file called <i>pcredemo.c</i> in the source
168    distribution. The
169    <a href="pcresample.html"><b>pcresample</b></a>
170    documentation describes how to run it.
171    </P>
172    <P>
173    A second matching function, <b>pcre_dfa_exec()</b>, which is not
174    Perl-compatible, is also provided. This uses a different algorithm for the
175    matching. The alternative algorithm finds all possible matches (at a given
176    point in the subject), and scans the subject just once. However, this algorithm
177    does not return captured substrings. A description of the two matching
178    algorithms and their advantages and disadvantages is given in the
179    <a href="pcrematching.html"><b>pcrematching</b></a>
180    documentation.
181  </P>  </P>
182  <P>  <P>
183    In addition to the main compiling and matching functions, there are convenience
184    functions for extracting captured substrings from a subject string that is
185    matched by <b>pcre_exec()</b>. They are:
186  <pre>  <pre>
187    <b>pcre_copy_substring()</b>    <b>pcre_copy_substring()</b>
188    <b>pcre_copy_named_substring()</b>    <b>pcre_copy_named_substring()</b>
189    <b>pcre_get_substring()</b>    <b>pcre_get_substring()</b>
190    <b>pcre_get_named_substring()</b>    <b>pcre_get_named_substring()</b>
191    <b>pcre_get_substring_list()</b>    <b>pcre_get_substring_list()</b>
192  </PRE>    <b>pcre_get_stringnumber()</b>
193  </P>    <b>pcre_get_stringtable_entries()</b>
194  <P>  </pre>
195  <b>pcre_free_substring()</b> and <b>pcre_free_substring_list()</b> are also  <b>pcre_free_substring()</b> and <b>pcre_free_substring_list()</b> are also
196  provided, to free the memory used for extracted strings.  provided, to free the memory used for extracted strings.
197  </P>  </P>
198  <P>  <P>
199  The function <b>pcre_maketables()</b> is used (optionally) to build a set of  The function <b>pcre_maketables()</b> is used to build a set of character tables
200  character tables in the current locale for passing to <b>pcre_compile()</b>.  in the current locale for passing to <b>pcre_compile()</b>, <b>pcre_exec()</b>,
201    or <b>pcre_dfa_exec()</b>. This is an optional facility that is provided for
202    specialist use. Most commonly, no special tables are passed, in which case
203    internal tables that are generated when PCRE is built are used.
204  </P>  </P>
205  <P>  <P>
206  The function <b>pcre_fullinfo()</b> is used to find out information about a  The function <b>pcre_fullinfo()</b> is used to find out information about a
207  compiled pattern; <b>pcre_info()</b> is an obsolete version which returns only  compiled pattern; <b>pcre_info()</b> is an obsolete version that returns only
208  some of the available information, but is retained for backwards compatibility.  some of the available information, but is retained for backwards compatibility.
209  The function <b>pcre_version()</b> returns a pointer to a string containing the  The function <b>pcre_version()</b> returns a pointer to a string containing the
210  version of PCRE and its date of release.  version of PCRE and its date of release.
211  </P>  </P>
212  <P>  <P>
213    The function <b>pcre_refcount()</b> maintains a reference count in a data block
214    containing a compiled pattern. This is provided for the benefit of
215    object-oriented applications.
216    </P>
217    <P>
218  The global variables <b>pcre_malloc</b> and <b>pcre_free</b> initially contain  The global variables <b>pcre_malloc</b> and <b>pcre_free</b> initially contain
219  the entry points of the standard <b>malloc()</b> and <b>free()</b> functions  the entry points of the standard <b>malloc()</b> and <b>free()</b> functions,
220  respectively. PCRE calls the memory management functions via these variables,  respectively. PCRE calls the memory management functions via these variables,
221  so a calling program can replace them if it wishes to intercept the calls. This  so a calling program can replace them if it wishes to intercept the calls. This
222  should be done before calling any PCRE functions.  should be done before calling any PCRE functions.
223  </P>  </P>
224  <P>  <P>
225    The global variables <b>pcre_stack_malloc</b> and <b>pcre_stack_free</b> are also
226    indirections to memory management functions. These special functions are used
227    only when PCRE is compiled to use the heap for remembering data, instead of
228    recursive function calls, when running the <b>pcre_exec()</b> function. See the
229    <a href="pcrebuild.html"><b>pcrebuild</b></a>
230    documentation for details of how to do this. It is a non-standard way of
231    building PCRE, for use in environments that have limited stacks. Because of the
232    greater use of memory management, it runs more slowly. Separate functions are
233    provided so that special-purpose external code can be used for this case. When
234    used, these functions are always called in a stack-like manner (last obtained,
235    first freed), and always for memory blocks of the same size. There is a
236    discussion about PCRE's stack usage in the
237    <a href="pcrestack.html"><b>pcrestack</b></a>
238    documentation.
239    </P>
240    <P>
241  The global variable <b>pcre_callout</b> initially contains NULL. It can be set  The global variable <b>pcre_callout</b> initially contains NULL. It can be set
242  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
243  points during a matching operation. Details are given in the <b>pcrecallout</b>  points during a matching operation. Details are given in the
244    <a href="pcrecallout.html"><b>pcrecallout</b></a>
245  documentation.  documentation.
246  </P>  </P>
247  <br><a name="SEC3" href="#TOC1">MULTITHREADING</a><br>  <br><a name="SEC3" href="#TOC1">NEWLINES</a><br>
248    <P>
249    PCRE supports five 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, any of the three preceding, or any
252    Unicode newline sequence. The Unicode newline sequences are the three just
253    mentioned, plus the single characters VT (vertical tab, U+000B), FF (formfeed,
254    U+000C), NEL (next line, U+0085), LS (line separator, U+2028), and PS
255    (paragraph separator, U+2029).
256    </P>
257    <P>
258    Each of the first three conventions is used by at least one operating system as
259    its standard newline sequence. When PCRE is built, a default can be specified.
260    The default default is LF, which is the Unix standard. When PCRE is run, the
261    default can be overridden, either when a pattern is compiled, or when it is
262    matched.
263    </P>
264    <P>
265    In the PCRE documentation the word "newline" is used to mean "the character or
266    pair of characters that indicate a line break". The choice of newline
267    convention affects the handling of the dot, circumflex, and dollar
268    metacharacters, the handling of #-comments in /x mode, and, when CRLF is a
269    recognized line ending sequence, the match position advancement for a
270    non-anchored pattern. The choice of newline convention does not affect the
271    interpretation of the \n or \r escape sequences.
272    </P>
273    <br><a name="SEC4" href="#TOC1">MULTITHREADING</a><br>
274  <P>  <P>
275  The PCRE functions can be used in multi-threading applications, with the  The PCRE functions can be used in multi-threading applications, with the
276  proviso that the memory management functions pointed to by <b>pcre_malloc</b>  proviso that the memory management functions pointed to by <b>pcre_malloc</b>,
277  and <b>pcre_free</b>, and the callout function pointed to by <b>pcre_callout</b>,  <b>pcre_free</b>, <b>pcre_stack_malloc</b>, and <b>pcre_stack_free</b>, and the
278  are shared by all threads.  callout function pointed to by <b>pcre_callout</b>, are shared by all threads.
279  </P>  </P>
280  <P>  <P>
281  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
282  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.
283  </P>  </P>
284  <br><a name="SEC4" href="#TOC1">CHECKING BUILD-TIME OPTIONS</a><br>  <br><a name="SEC5" href="#TOC1">SAVING PRECOMPILED PATTERNS FOR LATER USE</a><br>
285    <P>
286    The compiled form of a regular expression can be saved and re-used at a later
287    time, possibly by a different program, and even on a host other than the one on
288    which it was compiled. Details are given in the
289    <a href="pcreprecompile.html"><b>pcreprecompile</b></a>
290    documentation. However, compiling a regular expression with one version of PCRE
291    for use with a different version is not guaranteed to work and may cause
292    crashes.
293    </P>
294    <br><a name="SEC6" href="#TOC1">CHECKING BUILD-TIME OPTIONS</a><br>
295  <P>  <P>
296  <b>int pcre_config(int <i>what</i>, void *<i>where</i>);</b>  <b>int pcre_config(int <i>what</i>, void *<i>where</i>);</b>
297  </P>  </P>
# Line 186  documentation has more details about the Line 305  documentation has more details about the
305  The first argument for <b>pcre_config()</b> is an integer, specifying which  The first argument for <b>pcre_config()</b> is an integer, specifying which
306  information is required; the second argument is a pointer to a variable into  information is required; the second argument is a pointer to a variable into
307  which the information is placed. The following information is available:  which the information is placed. The following information is available:
 </P>  
 <P>  
308  <pre>  <pre>
309    PCRE_CONFIG_UTF8    PCRE_CONFIG_UTF8
310  </PRE>  </pre>
 </P>  
 <P>  
311  The output is an integer that is set to one if UTF-8 support is available;  The output is an integer that is set to one if UTF-8 support is available;
312  otherwise it is set to zero.  otherwise it is set to zero.
313  </P>  <pre>
314  <P>    PCRE_CONFIG_UNICODE_PROPERTIES
315    </pre>
316    The output is an integer that is set to one if support for Unicode character
317    properties is available; otherwise it is set to zero.
318  <pre>  <pre>
319    PCRE_CONFIG_NEWLINE    PCRE_CONFIG_NEWLINE
320  </PRE>  </pre>
321  </P>  The output is an integer whose value specifies the default character sequence
322  <P>  that is recognized as meaning "newline". The four values that are supported
323  The output is an integer that is set to the value of the code that is used for  are: 10 for LF, 13 for CR, 3338 for CRLF, -2 for ANYCRLF, and -1 for ANY. The
324  the newline character. It is either linefeed (10) or carriage return (13), and  default should normally be the standard sequence for your operating system.
 should normally be the standard character for your operating system.  
 </P>  
 <P>  
325  <pre>  <pre>
326    PCRE_CONFIG_LINK_SIZE    PCRE_CONFIG_LINK_SIZE
327  </PRE>  </pre>
 </P>  
 <P>  
328  The output is an integer that contains the number of bytes used for internal  The output is an integer that contains the number of bytes used for internal
329  linkage in compiled regular expressions. The value is 2, 3, or 4. Larger values  linkage in compiled regular expressions. The value is 2, 3, or 4. Larger values
330  allow larger regular expressions to be compiled, at the expense of slower  allow larger regular expressions to be compiled, at the expense of slower
331  matching. The default value of 2 is sufficient for all but the most massive  matching. The default value of 2 is sufficient for all but the most massive
332  patterns, since it allows the compiled pattern to be up to 64K in size.  patterns, since it allows the compiled pattern to be up to 64K in size.
 </P>  
 <P>  
333  <pre>  <pre>
334    PCRE_CONFIG_POSIX_MALLOC_THRESHOLD    PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
335  </PRE>  </pre>
 </P>  
 <P>  
336  The output is an integer that contains the threshold above which the POSIX  The output is an integer that contains the threshold above which the POSIX
337  interface uses <b>malloc()</b> for output vectors. Further details are given in  interface uses <b>malloc()</b> for output vectors. Further details are given in
338  the <b>pcreposix</b> documentation.  the
339  </P>  <a href="pcreposix.html"><b>pcreposix</b></a>
340  <P>  documentation.
341  <pre>  <pre>
342    PCRE_CONFIG_MATCH_LIMIT    PCRE_CONFIG_MATCH_LIMIT
343  </PRE>  </pre>
 </P>  
 <P>  
344  The output is an integer that gives the default limit for the number of  The output is an integer that gives the default limit for the number of
345  internal matching function calls in a <b>pcre_exec()</b> execution. Further  internal matching function calls in a <b>pcre_exec()</b> execution. Further
346  details are given with <b>pcre_exec()</b> below.  details are given with <b>pcre_exec()</b> below.
347    <pre>
348      PCRE_CONFIG_MATCH_LIMIT_RECURSION
349    </pre>
350    The output is an integer that gives the default limit for the depth of
351    recursion when calling the internal matching function in a <b>pcre_exec()</b>
352    execution. Further details are given with <b>pcre_exec()</b> below.
353    <pre>
354      PCRE_CONFIG_STACKRECURSE
355    </pre>
356    The output is an integer that is set to one if internal recursion when running
357    <b>pcre_exec()</b> is implemented by recursive function calls that use the stack
358    to remember their state. This is the usual way that PCRE is compiled. The
359    output is zero if PCRE was compiled to use blocks of data on the heap instead
360    of recursive function calls. In this case, <b>pcre_stack_malloc</b> and
361    <b>pcre_stack_free</b> are called to manage memory blocks on the heap, thus
362    avoiding the use of the stack.
363  </P>  </P>
364  <br><a name="SEC5" href="#TOC1">COMPILING A PATTERN</a><br>  <br><a name="SEC7" href="#TOC1">COMPILING A PATTERN</a><br>
365  <P>  <P>
366  <b>pcre *pcre_compile(const char *<i>pattern</i>, int <i>options</i>,</b>  <b>pcre *pcre_compile(const char *<i>pattern</i>, int <i>options</i>,</b>
367  <b>const char **<i>errptr</i>, int *<i>erroffset</i>,</b>  <b>const char **<i>errptr</i>, int *<i>erroffset</i>,</b>
368  <b>const unsigned char *<i>tableptr</i>);</b>  <b>const unsigned char *<i>tableptr</i>);</b>
369    <b>pcre *pcre_compile2(const char *<i>pattern</i>, int <i>options</i>,</b>
370    <b>int *<i>errorcodeptr</i>,</b>
371    <b>const char **<i>errptr</i>, int *<i>erroffset</i>,</b>
372    <b>const unsigned char *<i>tableptr</i>);</b>
373  </P>  </P>
374  <P>  <P>
375  The function <b>pcre_compile()</b> is called to compile a pattern into an  Either of the functions <b>pcre_compile()</b> or <b>pcre_compile2()</b> can be
376  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
377  is passed in the argument <i>pattern</i>. A pointer to a single block of memory  the two interfaces is that <b>pcre_compile2()</b> has an additional argument,
378  that is obtained via <b>pcre_malloc</b> is returned. This contains the compiled  <i>errorcodeptr</i>, via which a numerical error code can be returned.
379  code and related data. The <b>pcre</b> type is defined for the returned block;  </P>
380  this is a typedef for a structure whose contents are not externally defined. It  <P>
381  is up to the caller to free the memory when it is no longer required.  The pattern is a C string terminated by a binary zero, and is passed in the
382    <i>pattern</i> argument. A pointer to a single block of memory that is obtained
383    via <b>pcre_malloc</b> is returned. This contains the compiled code and related
384    data. The <b>pcre</b> type is defined for the returned block; this is a typedef
385    for a structure whose contents are not externally defined. It is up to the
386    caller to free the memory (via <b>pcre_free</b>) when it is no longer required.
387  </P>  </P>
388  <P>  <P>
389  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
390  depend on memory location, the complete <b>pcre</b> data block is not  depend on memory location, the complete <b>pcre</b> data block is not
391  fully relocatable, because it contains a copy of the <i>tableptr</i> argument,  fully relocatable, because it may contain a copy of the <i>tableptr</i>
392  which is an address (see below).  argument, which is an address (see below).
393  </P>  </P>
394  <P>  <P>
395  The <i>options</i> argument contains independent bits that affect the  The <i>options</i> argument contains various bit settings that affect the
396  compilation. It should be zero if no options are required. Some of the options,  compilation. It should be zero if no options are required. The available
397  in particular, those that are compatible with Perl, can also be set and unset  options are described below. Some of them, in particular, those that are
398  from within the pattern (see the detailed description of regular expressions  compatible with Perl, can also be set and unset from within the pattern (see
399  in the <b>pcrepattern</b> documentation). For these options, the contents of the  the detailed description in the
400  <i>options</i> argument specifies their initial settings at the start of  <a href="pcrepattern.html"><b>pcrepattern</b></a>
401  compilation and execution. The PCRE_ANCHORED option can be set at the time of  documentation). For these options, the contents of the <i>options</i> argument
402    specifies their initial settings at the start of compilation and execution. The
403    PCRE_ANCHORED and PCRE_NEWLINE_<i>xxx</i> options can be set at the time of
404  matching as well as at compile time.  matching as well as at compile time.
405  </P>  </P>
406  <P>  <P>
407  If <i>errptr</i> is NULL, <b>pcre_compile()</b> returns NULL immediately.  If <i>errptr</i> is NULL, <b>pcre_compile()</b> returns NULL immediately.
408  Otherwise, if compilation of a pattern fails, <b>pcre_compile()</b> returns  Otherwise, if compilation of a pattern fails, <b>pcre_compile()</b> returns
409  NULL, and sets the variable pointed to by <i>errptr</i> to point to a textual  NULL, and sets the variable pointed to by <i>errptr</i> to point to a textual
410  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
411  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
412    where the error was discovered is placed in the variable pointed to by
413  <i>erroffset</i>, which must not be NULL. If it is, an immediate error is given.  <i>erroffset</i>, which must not be NULL. If it is, an immediate error is given.
414  </P>  </P>
415  <P>  <P>
416  If the final argument, <i>tableptr</i>, is NULL, PCRE uses a default set of  If <b>pcre_compile2()</b> is used instead of <b>pcre_compile()</b>, and the
417  character tables which are built when it is compiled, using the default C  <i>errorcodeptr</i> argument is not NULL, a non-zero error code number is
418  locale. Otherwise, <i>tableptr</i> must be the result of a call to  returned via this argument in the event of an error. This is in addition to the
419  <b>pcre_maketables()</b>. See the section on locale support below.  textual error message. Error codes and messages are listed below.
420  </P>  </P>
421  <P>  <P>
422  This code fragment shows a typical straightforward call to <b>pcre_compile()</b>:  If the final argument, <i>tableptr</i>, is NULL, PCRE uses a default set of
423    character tables that are built when PCRE is compiled, using the default C
424    locale. Otherwise, <i>tableptr</i> must be an address that is the result of a
425    call to <b>pcre_maketables()</b>. This value is stored with the compiled
426    pattern, and used again by <b>pcre_exec()</b>, unless another table pointer is
427    passed to it. For more discussion, see the section on locale support below.
428  </P>  </P>
429  <P>  <P>
430    This code fragment shows a typical straightforward call to <b>pcre_compile()</b>:
431  <pre>  <pre>
432    pcre *re;    pcre *re;
433    const char *error;    const char *error;
# Line 297  This code fragment shows a typical strai Line 438  This code fragment shows a typical strai
438      &error,           /* for error message */      &error,           /* for error message */
439      &erroffset,       /* for error offset */      &erroffset,       /* for error offset */
440      NULL);            /* use default character tables */      NULL);            /* use default character tables */
441  </PRE>  </pre>
442  </P>  The following names for option bits are defined in the <b>pcre.h</b> header
443  <P>  file:
 The following option bits are defined:  
 </P>  
 <P>  
444  <pre>  <pre>
445    PCRE_ANCHORED    PCRE_ANCHORED
446  </PRE>  </pre>
 </P>  
 <P>  
447  If this bit is set, the pattern is forced to be "anchored", that is, it is  If this bit is set, the pattern is forced to be "anchored", that is, it is
448  constrained to match only at the first matching point in the string which is  constrained to match only at the first matching point in the string that is
449  being searched (the "subject string"). This effect can also be achieved by  being searched (the "subject string"). This effect can also be achieved by
450  appropriate constructs in the pattern itself, which is the only way to do it in  appropriate constructs in the pattern itself, which is the only way to do it in
451  Perl.  Perl.
452  </P>  <pre>
453  <P>    PCRE_AUTO_CALLOUT
454    </pre>
455    If this bit is set, <b>pcre_compile()</b> automatically inserts callout items,
456    all with number 255, before each pattern item. For discussion of the callout
457    facility, see the
458    <a href="pcrecallout.html"><b>pcrecallout</b></a>
459    documentation.
460  <pre>  <pre>
461    PCRE_CASELESS    PCRE_CASELESS
462  </PRE>  </pre>
 </P>  
 <P>  
463  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
464  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
465  pattern by a (?i) option setting.  pattern by a (?i) option setting. In UTF-8 mode, PCRE always understands the
466  </P>  concept of case for characters whose values are less than 128, so caseless
467  <P>  matching is always possible. For characters with higher values, the concept of
468    case is supported if PCRE is compiled with Unicode property support, but not
469    otherwise. If you want to use caseless matching for characters 128 and above,
470    you must ensure that PCRE is compiled with Unicode property support as well as
471    with UTF-8 support.
472  <pre>  <pre>
473    PCRE_DOLLAR_ENDONLY    PCRE_DOLLAR_ENDONLY
474  </PRE>  </pre>
 </P>  
 <P>  
475  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
476  end of the subject string. Without this option, a dollar also matches  end of the subject string. Without this option, a dollar also matches
477  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
478  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.
479  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
480  a pattern.  pattern.
 </P>  
 <P>  
481  <pre>  <pre>
482    PCRE_DOTALL    PCRE_DOTALL
483  </PRE>  </pre>
 </P>  
 <P>  
484  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,
485  including newlines. Without it, newlines are excluded. This option is  including those that indicate newline. Without it, a dot does not match when
486  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
487  (?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
488  character, independent of the setting of this option.  negative class such as [^a] always matches newline characters, independent of
489  </P>  the setting of this option.
490  <P>  <pre>
491      PCRE_DUPNAMES
492    </pre>
493    If this bit is set, names used to identify capturing subpatterns need not be
494    unique. This can be helpful for certain types of pattern when it is known that
495    only one instance of the named subpattern can ever be matched. There are more
496    details of named subpatterns below; see also the
497    <a href="pcrepattern.html"><b>pcrepattern</b></a>
498    documentation.
499  <pre>  <pre>
500    PCRE_EXTENDED    PCRE_EXTENDED
501  </PRE>  </pre>
 </P>  
 <P>  
502  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
503  ignored except when escaped or inside a character class. Whitespace does not  ignored except when escaped or inside a character class. Whitespace does not
504  include the VT character (code 11). In addition, characters between an  include the VT character (code 11). In addition, characters between an
505  unescaped # outside a character class and the next newline character,  unescaped # outside a character class and the next newline, inclusive, are also
506  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
507  be changed within a pattern by a (?x) option setting.  pattern by a (?x) option setting.
508  </P>  </P>
509  <P>  <P>
510  This option makes it possible to include comments inside complicated patterns.  This option makes it possible to include comments inside complicated patterns.
511  Note, however, that this applies only to data characters. Whitespace characters  Note, however, that this applies only to data characters. Whitespace characters
512  may never appear within special character sequences in a pattern, for example  may never appear within special character sequences in a pattern, for example
513  within the sequence (?( which introduces a conditional subpattern.  within the sequence (?( which introduces a conditional subpattern.
 </P>  
 <P>  
514  <pre>  <pre>
515    PCRE_EXTRA    PCRE_EXTRA
516  </PRE>  </pre>
 </P>  
 <P>  
517  This option was invented in order to turn on additional functionality of PCRE  This option was invented in order to turn on additional functionality of PCRE
518  that is incompatible with Perl, but it is currently of very little use. When  that is incompatible with Perl, but it is currently of very little use. When
519  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
520  special meaning causes an error, thus reserving these combinations for future  special meaning causes an error, thus reserving these combinations for future
521  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
522  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
523  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
524  pattern.  this option. It can also be set by a (?X) option setting within a pattern.
525  </P>  <pre>
526  <P>    PCRE_FIRSTLINE
527    </pre>
528    If this option is set, an unanchored pattern is required to match before or at
529    the first newline in the subject string, though the matched text may continue
530    over the newline.
531  <pre>  <pre>
532    PCRE_MULTILINE    PCRE_MULTILINE
533  </PRE>  </pre>
534  </P>  By default, PCRE treats the subject string as consisting of a single line of
535  <P>  characters (even if it actually contains newlines). The "start of line"
 By default, PCRE treats the subject string as consisting of a single "line" of  
 characters (even if it actually contains several newlines). The "start of line"  
536  metacharacter (^) matches only at the start of the string, while the "end of  metacharacter (^) matches only at the start of the string, while the "end of
537  line" metacharacter ($) matches only at the end of the string, or before a  line" metacharacter ($) matches only at the end of the string, or before a
538  terminating newline (unless PCRE_DOLLAR_ENDONLY is set). This is the same as  terminating newline (unless PCRE_DOLLAR_ENDONLY is set). This is the same as
# Line 398  Perl. Line 540  Perl.
540  </P>  </P>
541  <P>  <P>
542  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
543  match immediately following or immediately before any newline in the subject  match immediately following or immediately before internal newlines in the
544  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
545  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
546  setting. If there are no "\n" characters in a subject string, or no  (?m) option setting. If there are no newlines in a subject string, or no
547  occurrences of ^ or $ in a pattern, setting PCRE_MULTILINE has no effect.  occurrences of ^ or $ in a pattern, setting PCRE_MULTILINE has no effect.
548    <pre>
549      PCRE_NEWLINE_CR
550      PCRE_NEWLINE_LF
551      PCRE_NEWLINE_CRLF
552      PCRE_NEWLINE_ANYCRLF
553      PCRE_NEWLINE_ANY
554    </pre>
555    These options override the default newline definition that was chosen when PCRE
556    was built. Setting the first or the second specifies that a newline is
557    indicated by a single character (CR or LF, respectively). Setting
558    PCRE_NEWLINE_CRLF specifies that a newline is indicated by the two-character
559    CRLF sequence. Setting PCRE_NEWLINE_ANYCRLF specifies that any of the three
560    preceding sequences should be recognized. Setting PCRE_NEWLINE_ANY specifies
561    that any Unicode newline sequence should be recognized. The Unicode newline
562    sequences are the three just mentioned, plus the single characters VT (vertical
563    tab, U+000B), FF (formfeed, U+000C), NEL (next line, U+0085), LS (line
564    separator, U+2028), and PS (paragraph separator, U+2029). The last two are
565    recognized only in UTF-8 mode.
566    </P>
567    <P>
568    The newline setting in the options word uses three bits that are treated
569    as a number, giving eight possibilities. Currently only six are used (default
570    plus the five values above). This means that if you set more than one newline
571    option, the combination may or may not be sensible. For example,
572    PCRE_NEWLINE_CR with PCRE_NEWLINE_LF is equivalent to PCRE_NEWLINE_CRLF, but
573    other combinations may yield unused numbers and cause an error.
574    </P>
575    <P>
576    The only time that a line break is specially recognized when compiling a
577    pattern is if PCRE_EXTENDED is set, and an unescaped # outside a character
578    class is encountered. This indicates a comment that lasts until after the next
579    line break sequence. In other circumstances, line break sequences are treated
580    as literal data, except that in PCRE_EXTENDED mode, both CR and LF are treated
581    as whitespace characters and are therefore ignored.
582  </P>  </P>
583  <P>  <P>
584    The newline option that is set at compile time becomes the default that is used
585    for <b>pcre_exec()</b> and <b>pcre_dfa_exec()</b>, but it can be overridden.
586  <pre>  <pre>
587    PCRE_NO_AUTO_CAPTURE    PCRE_NO_AUTO_CAPTURE
588  </PRE>  </pre>
 </P>  
 <P>  
589  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
590  the pattern. Any opening parenthesis that is not followed by ? behaves as if it  the pattern. Any opening parenthesis that is not followed by ? behaves as if it
591  were followed by ?: but named parentheses can still be used for capturing (and  were followed by ?: but named parentheses can still be used for capturing (and
592  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
593  in Perl.  in Perl.
 </P>  
 <P>  
594  <pre>  <pre>
595    PCRE_UNGREEDY    PCRE_UNGREEDY
596  </PRE>  </pre>
 </P>  
 <P>  
597  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
598  greedy by default, but become greedy if followed by "?". It is not compatible  greedy by default, but become greedy if followed by "?". It is not compatible
599  with Perl. It can also be set by a (?U) option setting within the pattern.  with Perl. It can also be set by a (?U) option setting within the pattern.
 </P>  
 <P>  
600  <pre>  <pre>
601    PCRE_UTF8    PCRE_UTF8
602  </PRE>  </pre>
 </P>  
 <P>  
603  This option causes PCRE to regard both the pattern and the subject as strings  This option causes PCRE to regard both the pattern and the subject as strings
604  of UTF-8 characters instead of single-byte character strings. However, it is  of UTF-8 characters instead of single-byte character strings. However, it is
605  available only if PCRE has been 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
606  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
607  behaviour of PCRE are given in the  behaviour of PCRE are given in the
608  <a href="pcre.html#utf8support">section on UTF-8 support</a>  <a href="pcre.html#utf8support">section on UTF-8 support</a>
609  in the main  in the main
610  <a href="pcre.html"><b>pcre</b></a>  <a href="pcre.html"><b>pcre</b></a>
611  page.  page.
612    <pre>
613      PCRE_NO_UTF8_CHECK
614    </pre>
615    When PCRE_UTF8 is set, the validity of the pattern as a UTF-8 string is
616    automatically checked. If an invalid UTF-8 sequence of bytes is found,
617    <b>pcre_compile()</b> returns an error. If you already know that your pattern is
618    valid, and you want to skip this check for performance reasons, you can set the
619    PCRE_NO_UTF8_CHECK option. When it is set, the effect of passing an invalid
620    UTF-8 string as a pattern is undefined. It may cause your program to crash.
621    Note that this option can also be passed to <b>pcre_exec()</b> and
622    <b>pcre_dfa_exec()</b>, to suppress the UTF-8 validity checking of subject
623    strings.
624    </P>
625    <br><a name="SEC8" href="#TOC1">COMPILATION ERROR CODES</a><br>
626    <P>
627    The following table lists the error codes than may be returned by
628    <b>pcre_compile2()</b>, along with the error messages that may be returned by
629    both compiling functions. As PCRE has developed, some error codes have fallen
630    out of use. To avoid confusion, they have not been re-used.
631    <pre>
632       0  no error
633       1  \ at end of pattern
634       2  \c at end of pattern
635       3  unrecognized character follows \
636       4  numbers out of order in {} quantifier
637       5  number too big in {} quantifier
638       6  missing terminating ] for character class
639       7  invalid escape sequence in character class
640       8  range out of order in character class
641       9  nothing to repeat
642      10  [this code is not in use]
643      11  internal error: unexpected repeat
644      12  unrecognized character after (?
645      13  POSIX named classes are supported only within a class
646      14  missing )
647      15  reference to non-existent subpattern
648      16  erroffset passed as NULL
649      17  unknown option bit(s) set
650      18  missing ) after comment
651      19  [this code is not in use]
652      20  regular expression too large
653      21  failed to get memory
654      22  unmatched parentheses
655      23  internal error: code overflow
656      24  unrecognized character after (?&#60;
657      25  lookbehind assertion is not fixed length
658      26  malformed number or name after (?(
659      27  conditional group contains more than two branches
660      28  assertion expected after (?(
661      29  (?R or (?digits must be followed by )
662      30  unknown POSIX class name
663      31  POSIX collating elements are not supported
664      32  this version of PCRE is not compiled with PCRE_UTF8 support
665      33  [this code is not in use]
666      34  character value in \x{...} sequence is too large
667      35  invalid condition (?(0)
668      36  \C not allowed in lookbehind assertion
669      37  PCRE does not support \L, \l, \N, \U, or \u
670      38  number after (?C is &#62; 255
671      39  closing ) for (?C expected
672      40  recursive call could loop indefinitely
673      41  unrecognized character after (?P
674      42  syntax error in subpattern name (missing terminator)
675      43  two named subpatterns have the same name
676      44  invalid UTF-8 string
677      45  support for \P, \p, and \X has not been compiled
678      46  malformed \P or \p sequence
679      47  unknown property name after \P or \p
680      48  subpattern name is too long (maximum 32 characters)
681      49  too many named subpatterns (maximum 10,000)
682      50  repeated subpattern is too long
683      51  octal value is greater than \377 (not in UTF-8 mode)
684      52  internal error: overran compiling workspace
685      53  internal error: previously-checked referenced subpattern not found
686      54  DEFINE group contains more than one branch
687      55  repeating a DEFINE group is not allowed
688      56  inconsistent NEWLINE options"
689    </PRE>
690  </P>  </P>
691  <br><a name="SEC6" href="#TOC1">STUDYING A PATTERN</a><br>  <br><a name="SEC9" href="#TOC1">STUDYING A PATTERN</a><br>
692  <P>  <P>
693  <b>pcre_extra *pcre_study(const pcre *<i>code</i>, int <i>options</i>,</b>  <b>pcre_extra *pcre_study(const pcre *<i>code</i>, int <i>options</i></b>
694  <b>const char **<i>errptr</i>);</b>  <b>const char **<i>errptr</i>);</b>
695  </P>  </P>
696  <P>  <P>
697  When a pattern is going to be used several times, it is worth spending more  If a compiled pattern is going to be used several times, it is worth spending
698  time analyzing it in order to speed up the time taken for matching. The  more time analyzing it in order to speed up the time taken for matching. The
699  function <b>pcre_study()</b> takes a pointer to a compiled pattern as its first  function <b>pcre_study()</b> takes a pointer to a compiled pattern as its first
700  argument. If studing the pattern produces additional information that will help  argument. If studying the pattern produces additional information that will
701  speed up matching, <b>pcre_study()</b> returns a pointer to a <b>pcre_extra</b>  help speed up matching, <b>pcre_study()</b> returns a pointer to a
702  block, in which the <i>study_data</i> field points to the results of the study.  <b>pcre_extra</b> block, in which the <i>study_data</i> field points to the
703    results of the study.
704  </P>  </P>
705  <P>  <P>
706  The returned value from a <b>pcre_study()</b> can be passed directly to  The returned value from <b>pcre_study()</b> can be passed directly to
707  <b>pcre_exec()</b>. However, the <b>pcre_extra</b> block also contains other  <b>pcre_exec()</b>. However, a <b>pcre_extra</b> block also contains other
708  fields that can be set by the caller before the block is passed; these are  fields that can be set by the caller before the block is passed; these are
709  described below. If studying the pattern does not produce any additional  described
710  information, <b>pcre_study()</b> returns NULL. In that circumstance, if the  <a href="#extradata">below</a>
711  calling program wants to pass some of the other fields to <b>pcre_exec()</b>, it  in the section on matching a pattern.
 must set up its own <b>pcre_extra</b> block.  
712  </P>  </P>
713  <P>  <P>
714  The second argument contains option bits. At present, no options are defined  If studying the pattern does not produce any additional information
715  for <b>pcre_study()</b>, and this argument should always be zero.  <b>pcre_study()</b> returns NULL. In that circumstance, if the calling program
716    wants to pass any of the other fields to <b>pcre_exec()</b>, it must set up its
717    own <b>pcre_extra</b> block.
718    </P>
719    <P>
720    The second argument of <b>pcre_study()</b> contains option bits. At present, no
721    options are defined, and this argument should always be zero.
722  </P>  </P>
723  <P>  <P>
724  The third argument for <b>pcre_study()</b> is a pointer for an error message. If  The third argument for <b>pcre_study()</b> is a pointer for an error message. If
725  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
726  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
727  therefore test the error pointer for NULL after calling <b>pcre_study()</b>, to  static string that is part of the library. You must not try to free it. You
728  be sure that it has run successfully.  should test the error pointer for NULL after calling <b>pcre_study()</b>, to be
729    sure that it has run successfully.
730  </P>  </P>
731  <P>  <P>
732  This is a typical call to <b>pcre_study</b>():  This is a typical call to <b>pcre_study</b>():
 </P>  
 <P>  
733  <pre>  <pre>
734    pcre_extra *pe;    pcre_extra *pe;
735    pe = pcre_study(    pe = pcre_study(
736      re,             /* result of pcre_compile() */      re,             /* result of pcre_compile() */
737      0,              /* no options exist */      0,              /* no options exist */
738      &error);        /* set to NULL or points to a message */      &error);        /* set to NULL or points to a message */
739  </PRE>  </pre>
 </P>  
 <P>  
740  At present, studying a pattern is useful only for non-anchored patterns that do  At present, studying a pattern is useful only for non-anchored patterns that do
741  not have a single fixed starting character. A bitmap of possible starting  not have a single fixed starting character. A bitmap of possible starting
742  characters is created.  bytes is created.
743  </P>  <a name="localesupport"></a></P>
744  <a name="localesupport"></a><br><a name="SEC7" href="#TOC1">LOCALE SUPPORT</a><br>  <br><a name="SEC10" href="#TOC1">LOCALE SUPPORT</a><br>
745  <P>  <P>
746  PCRE handles caseless matching, and determines whether characters are letters,  PCRE handles caseless matching, and determines whether characters are letters,
747  digits, or whatever, by reference to a set of tables. When running in UTF-8  digits, or whatever, by reference to a set of tables, indexed by character
748  mode, this applies only to characters with codes less than 256. The library  value. When running in UTF-8 mode, this applies only to characters with codes
749  contains a default set of tables that is created in the default C locale when  less than 128. Higher-valued codes never match escapes such as \w or \d, but
750  PCRE is compiled. This is used when the final argument of <b>pcre_compile()</b>  can be tested with \p if PCRE is built with Unicode character property
751  is NULL, and is sufficient for many applications.  support. The use of locales with Unicode is discouraged. If you are handling
752  </P>  characters with codes greater than 128, you should either use UTF-8 and
753  <P>  Unicode, or use locales, but not try to mix the two.
754  An alternative set of tables can, however, be supplied. Such tables are built  </P>
755  by calling the <b>pcre_maketables()</b> function, which has no arguments, in the  <P>
756  relevant locale. The result can then be passed to <b>pcre_compile()</b> as often  PCRE contains an internal set of tables that are used when the final argument
757  as necessary. For example, to build and use tables that are appropriate for the  of <b>pcre_compile()</b> is NULL. These are sufficient for many applications.
758  French locale (where accented characters with codes greater than 128 are  Normally, the internal tables recognize only ASCII characters. However, when
759  treated as letters), the following code could be used:  PCRE is built, it is possible to cause the internal tables to be rebuilt in the
760  </P>  default "C" locale of the local system, which may cause them to be different.
761  <P>  </P>
762    <P>
763    The internal tables can always be overridden by tables supplied by the
764    application that calls PCRE. These may be created in a different locale from
765    the default. As more and more applications change to using Unicode, the need
766    for this locale support is expected to die away.
767    </P>
768    <P>
769    External tables are built by calling the <b>pcre_maketables()</b> function,
770    which has no arguments, in the relevant locale. The result can then be passed
771    to <b>pcre_compile()</b> or <b>pcre_exec()</b> as often as necessary. For
772    example, to build and use tables that are appropriate for the French locale
773    (where accented characters with values greater than 128 are treated as letters),
774    the following code could be used:
775  <pre>  <pre>
776    setlocale(LC_CTYPE, "fr");    setlocale(LC_CTYPE, "fr_FR");
777    tables = pcre_maketables();    tables = pcre_maketables();
778    re = pcre_compile(..., tables);    re = pcre_compile(..., tables);
779  </PRE>  </pre>
780    The locale name "fr_FR" is used on Linux and other Unix-like systems; if you
781    are using Windows, the name for the French locale is "french".
782  </P>  </P>
783  <P>  <P>
784  The tables are built in memory that is obtained via <b>pcre_malloc</b>. The  When <b>pcre_maketables()</b> runs, the tables are built in memory that is
785  pointer that is passed to <b>pcre_compile</b> is saved with the compiled  obtained via <b>pcre_malloc</b>. It is the caller's responsibility to ensure
786    that the memory containing the tables remains available for as long as it is
787    needed.
788    </P>
789    <P>
790    The pointer that is passed to <b>pcre_compile()</b> is saved with the compiled
791  pattern, and the same tables are used via this pointer by <b>pcre_study()</b>  pattern, and the same tables are used via this pointer by <b>pcre_study()</b>
792  and <b>pcre_exec()</b>. Thus, for any single pattern, compilation, studying and  and normally also by <b>pcre_exec()</b>. Thus, by default, for any single
793  matching all happen in the same locale, but different patterns can be compiled  pattern, compilation, studying and matching all happen in the same locale, but
794  in different locales. It is the caller's responsibility to ensure that the  different patterns can be compiled in different locales.
795  memory containing the tables remains available for as long as it is needed.  </P>
796    <P>
797    It is possible to pass a table pointer or NULL (indicating the use of the
798    internal tables) to <b>pcre_exec()</b>. Although not intended for this purpose,
799    this facility could be used to match a pattern in a different locale from the
800    one in which it was compiled. Passing table pointers at run time is discussed
801    below in the section on matching a pattern.
802  </P>  </P>
803  <br><a name="SEC8" href="#TOC1">INFORMATION ABOUT A PATTERN</a><br>  <br><a name="SEC11" href="#TOC1">INFORMATION ABOUT A PATTERN</a><br>
804  <P>  <P>
805  <b>int pcre_fullinfo(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>  <b>int pcre_fullinfo(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>
806  <b>int <i>what</i>, void *<i>where</i>);</b>  <b>int <i>what</i>, void *<i>where</i>);</b>
# Line 542  the pattern was not studied. The third a Line 817  the pattern was not studied. The third a
817  information is required, and the fourth argument is a pointer to a variable  information is required, and the fourth argument is a pointer to a variable
818  to receive the data. The yield of the function is zero for success, or one of  to receive the data. The yield of the function is zero for success, or one of
819  the following negative numbers:  the following negative numbers:
 </P>  
 <P>  
820  <pre>  <pre>
821    PCRE_ERROR_NULL       the argument <i>code</i> was NULL    PCRE_ERROR_NULL       the argument <i>code</i> was NULL
822                          the argument <i>where</i> was NULL                          the argument <i>where</i> was NULL
823    PCRE_ERROR_BADMAGIC   the "magic number" was not found    PCRE_ERROR_BADMAGIC   the "magic number" was not found
824    PCRE_ERROR_BADOPTION  the value of <i>what</i> was invalid    PCRE_ERROR_BADOPTION  the value of <i>what</i> was invalid
825  </PRE>  </pre>
826  </P>  The "magic number" is placed at the start of each compiled pattern as an simple
827  <P>  check against passing an arbitrary memory pointer. Here is a typical call of
828  Here is a typical call of <b>pcre_fullinfo()</b>, to obtain the length of the  <b>pcre_fullinfo()</b>, to obtain the length of the compiled pattern:
 compiled pattern:  
 </P>  
 <P>  
829  <pre>  <pre>
830    int rc;    int rc;
831    unsigned long int length;    size_t length;
832    rc = pcre_fullinfo(    rc = pcre_fullinfo(
833      re,               /* result of pcre_compile() */      re,               /* result of pcre_compile() */
834      pe,               /* result of pcre_study(), or NULL */      pe,               /* result of pcre_study(), or NULL */
835      PCRE_INFO_SIZE,   /* what is required */      PCRE_INFO_SIZE,   /* what is required */
836      &length);         /* where to put the data */      &length);         /* where to put the data */
837  </PRE>  </pre>
 </P>  
 <P>  
838  The possible values for the third argument are defined in <b>pcre.h</b>, and are  The possible values for the third argument are defined in <b>pcre.h</b>, and are
839  as follows:  as follows:
 </P>  
 <P>  
840  <pre>  <pre>
841    PCRE_INFO_BACKREFMAX    PCRE_INFO_BACKREFMAX
842  </PRE>  </pre>
 </P>  
 <P>  
843  Return the number of the highest back reference in the pattern. The fourth  Return the number of the highest back reference in the pattern. The fourth
844  argument should point to an <b>int</b> variable. Zero is returned if there are  argument should point to an <b>int</b> variable. Zero is returned if there are
845  no back references.  no back references.
 </P>  
 <P>  
846  <pre>  <pre>
847    PCRE_INFO_CAPTURECOUNT    PCRE_INFO_CAPTURECOUNT
848  </PRE>  </pre>
 </P>  
 <P>  
849  Return the number of capturing subpatterns in the pattern. The fourth argument  Return the number of capturing subpatterns in the pattern. The fourth argument
850  should point to an \fbint\fR variable.  should point to an <b>int</b> variable.
851  </P>  <pre>
852  <P>    PCRE_INFO_DEFAULT_TABLES
853    </pre>
854    Return a pointer to the internal default character tables within PCRE. The
855    fourth argument should point to an <b>unsigned char *</b> variable. This
856    information call is provided for internal use by the <b>pcre_study()</b>
857    function. External callers can cause PCRE to use its internal tables by passing
858    a NULL table pointer.
859  <pre>  <pre>
860    PCRE_INFO_FIRSTBYTE    PCRE_INFO_FIRSTBYTE
861  </PRE>  </pre>
 </P>  
 <P>  
862  Return information about the first byte of any matched string, for a  Return information about the first byte of any matched string, for a
863  non-anchored pattern. (This option used to be called PCRE_INFO_FIRSTCHAR; the  non-anchored pattern. The fourth argument should point to an <b>int</b>
864  old name is still recognized for backwards compatibility.)  variable. (This option used to be called PCRE_INFO_FIRSTCHAR; the old name is
865  </P>  still recognized for backwards compatibility.)
 <P>  
 If there is a fixed first byte, e.g. from a pattern such as (cat|cow|coyote),  
 it is returned in the integer pointed to by <i>where</i>. Otherwise, if either  
866  </P>  </P>
867  <P>  <P>
868    If there is a fixed first byte, for example, from a pattern such as
869    (cat|cow|coyote), its value is returned. Otherwise, if either
870    <br>
871    <br>
872  (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
873  starts with "^", or  starts with "^", or
874  </P>  <br>
875  <P>  <br>
876  (b) every branch of the pattern starts with ".*" and PCRE_DOTALL is not set  (b) every branch of the pattern starts with ".*" and PCRE_DOTALL is not set
877  (if it were set, the pattern would be anchored),  (if it were set, the pattern would be anchored),
878  </P>  <br>
879  <P>  <br>
880  -1 is returned, indicating that the pattern matches only at the start of a  -1 is returned, indicating that the pattern matches only at the start of a
881  subject string or after any newline within the string. Otherwise -2 is  subject string or after any newline within the string. Otherwise -2 is
882  returned. For anchored patterns, -2 is returned.  returned. For anchored patterns, -2 is returned.
 </P>  
 <P>  
883  <pre>  <pre>
884    PCRE_INFO_FIRSTTABLE    PCRE_INFO_FIRSTTABLE
885  </PRE>  </pre>
 </P>  
 <P>  
886  If the pattern was studied, and this resulted in the construction of a 256-bit  If the pattern was studied, and this resulted in the construction of a 256-bit
887  table indicating a fixed set of bytes for the first byte in any matching  table indicating a fixed set of bytes for the first byte in any matching
888  string, a pointer to the table is returned. Otherwise NULL is returned. The  string, a pointer to the table is returned. Otherwise NULL is returned. The
889  fourth argument should point to an <b>unsigned char *</b> variable.  fourth argument should point to an <b>unsigned char *</b> variable.
890  </P>  <pre>
891  <P>    PCRE_INFO_JCHANGED
892    </pre>
893    Return 1 if the (?J) option setting is used in the pattern, otherwise 0. The
894    fourth argument should point to an <b>int</b> variable. The (?J) internal option
895    setting changes the local PCRE_DUPNAMES value.
896  <pre>  <pre>
897    PCRE_INFO_LASTLITERAL    PCRE_INFO_LASTLITERAL
898  </PRE>  </pre>
 </P>  
 <P>  
899  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
900  string, other than at its start, if such a byte has been recorded. The fourth  string, other than at its start, if such a byte has been recorded. The fourth
901  argument should point to an <b>int</b> variable. If there is no such byte, -1 is  argument should point to an <b>int</b> variable. If there is no such byte, -1 is
# Line 640  returned. For anchored patterns, a last Line 903  returned. For anchored patterns, a last
903  follows something of variable length. For example, for the pattern  follows something of variable length. For example, for the pattern
904  /^a\d+z\d+/ the returned value is "z", but for /^a\dz\d/ the returned value  /^a\d+z\d+/ the returned value is "z", but for /^a\dz\d/ the returned value
905  is -1.  is -1.
 </P>  
 <P>  
906  <pre>  <pre>
907    PCRE_INFO_NAMECOUNT    PCRE_INFO_NAMECOUNT
908    PCRE_INFO_NAMEENTRYSIZE    PCRE_INFO_NAMEENTRYSIZE
909    PCRE_INFO_NAMETABLE    PCRE_INFO_NAMETABLE
910  </PRE>  </pre>
 </P>  
 <P>  
911  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
912  names are just an additional way of identifying the parentheses, which still  names are just an additional way of identifying the parentheses, which still
913  acquire a number. A caller that wants to extract data from a named subpattern  acquire numbers. Several convenience functions such as
914  must convert the name to a number in order to access the correct pointers in  <b>pcre_get_named_substring()</b> are provided for extracting captured
915  the output vector (described with <b>pcre_exec()</b> below). In order to do  substrings by name. It is also possible to extract the data directly, by first
916  this, it must first use these three values to obtain the name-to-number mapping  converting the name to a number in order to access the correct pointers in the
917  table for the pattern.  output vector (described with <b>pcre_exec()</b> below). To do the conversion,
918    you need to use the name-to-number map, which is described by these three
919    values.
920  </P>  </P>
921  <P>  <P>
922  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
# Line 665  length of the longest name. PCRE_INFO_NA Line 926  length of the longest name. PCRE_INFO_NA
926  entry of the table (a pointer to <b>char</b>). The first two bytes of each entry  entry of the table (a pointer to <b>char</b>). The first two bytes of each entry
927  are the number of the capturing parenthesis, most significant byte first. The  are the number of the capturing parenthesis, most significant byte first. The
928  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
929  alphabetical order. For example, consider the following pattern (assume  alphabetical order. When PCRE_DUPNAMES is set, duplicate names are in order of
930    their parentheses numbers. For example, consider the following pattern (assume
931  PCRE_EXTENDED is set, so white space - including newlines - is ignored):  PCRE_EXTENDED is set, so white space - including newlines - is ignored):
 </P>  
 <P>  
932  <pre>  <pre>
933    (?P&#60;date&#62; (?P&#60;year&#62;(\d\d)?\d\d) -    (?&#60;date&#62; (?&#60;year&#62;(\d\d)?\d\d) - (?&#60;month&#62;\d\d) - (?&#60;day&#62;\d\d) )
934    (?P&#60;month&#62;\d\d) - (?P&#60;day&#62;\d\d) )  </pre>
 </PRE>  
 </P>  
 <P>  
935  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
936  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
937  bytes shows in hex, and undefined bytes shown as ??:  bytes shows in hexadecimal, and undefined bytes shown as ??:
 </P>  
 <P>  
938  <pre>  <pre>
939    00 01 d  a  t  e  00 ??    00 01 d  a  t  e  00 ??
940    00 05 d  a  y  00 ?? ??    00 05 d  a  y  00 ?? ??
941    00 04 m  o  n  t  h  00    00 04 m  o  n  t  h  00
942    00 02 y  e  a  r  00 ??    00 02 y  e  a  r  00 ??
943  </PRE>  </pre>
944  </P>  When writing code to extract data from named subpatterns using the
945  <P>  name-to-number map, remember that the length of the entries is likely to be
946  When writing code to extract data from named subpatterns, remember that the  different for each compiled pattern.
947  length of each entry may be different for each compiled pattern.  <pre>
948  </P>    PCRE_INFO_OKPARTIAL
949  <P>  </pre>
950    Return 1 if the pattern can be used for partial matching, otherwise 0. The
951    fourth argument should point to an <b>int</b> variable. The
952    <a href="pcrepartial.html"><b>pcrepartial</b></a>
953    documentation lists the restrictions that apply to patterns when partial
954    matching is used.
955  <pre>  <pre>
956    PCRE_INFO_OPTIONS    PCRE_INFO_OPTIONS
957  </PRE>  </pre>
 </P>  
 <P>  
958  Return a copy of the options with which the pattern was compiled. The fourth  Return a copy of the options with which the pattern was compiled. The fourth
959  argument should point to an <b>unsigned long int</b> variable. These option bits  argument should point to an <b>unsigned long int</b> variable. These option bits
960  are those specified in the call to <b>pcre_compile()</b>, modified by any  are those specified in the call to <b>pcre_compile()</b>, modified by any
# Line 705  top-level option settings within the pat Line 963  top-level option settings within the pat
963  <P>  <P>
964  A pattern is automatically anchored by PCRE if all of its top-level  A pattern is automatically anchored by PCRE if all of its top-level
965  alternatives begin with one of the following:  alternatives begin with one of the following:
 </P>  
 <P>  
966  <pre>  <pre>
967    ^     unless PCRE_MULTILINE is set    ^     unless PCRE_MULTILINE is set
968    \A    always    \A    always
969    \G    always    \G    always
970    .*    if PCRE_DOTALL is set and there are no back    .*    if PCRE_DOTALL is set and there are no back references to the subpattern in which .* appears
971            references to the subpattern in which .* appears  </pre>
 </PRE>  
 </P>  
 <P>  
972  For such patterns, the PCRE_ANCHORED bit is set in the options returned by  For such patterns, the PCRE_ANCHORED bit is set in the options returned by
973  <b>pcre_fullinfo()</b>.  <b>pcre_fullinfo()</b>.
 </P>  
 <P>  
974  <pre>  <pre>
975    PCRE_INFO_SIZE    PCRE_INFO_SIZE
976  </PRE>  </pre>
 </P>  
 <P>  
977  Return the size of the compiled pattern, that is, the value that was passed as  Return the size of the compiled pattern, that is, the value that was passed as
978  the argument to <b>pcre_malloc()</b> when PCRE was getting memory in which to  the argument to <b>pcre_malloc()</b> when PCRE was getting memory in which to
979  place the compiled data. The fourth argument should point to a <b>size_t</b>  place the compiled data. The fourth argument should point to a <b>size_t</b>
980  variable.  variable.
 </P>  
 <P>  
981  <pre>  <pre>
982    PCRE_INFO_STUDYSIZE    PCRE_INFO_STUDYSIZE
983  </PRE>  </pre>
984  </P>  Return the size of the data block pointed to by the <i>study_data</i> field in
 <P>  
 Returns the size of the data block pointed to by the <i>study_data</i> field in  
985  a <b>pcre_extra</b> block. That is, it is the value that was passed to  a <b>pcre_extra</b> block. That is, it is the value that was passed to
986  <b>pcre_malloc()</b> when PCRE was getting memory into which to place the data  <b>pcre_malloc()</b> when PCRE was getting memory into which to place the data
987  created by <b>pcre_study()</b>. The fourth argument should point to a  created by <b>pcre_study()</b>. The fourth argument should point to a
988  <b>size_t</b> variable.  <b>size_t</b> variable.
989  </P>  </P>
990  <br><a name="SEC9" href="#TOC1">OBSOLETE INFO FUNCTION</a><br>  <br><a name="SEC12" href="#TOC1">OBSOLETE INFO FUNCTION</a><br>
991  <P>  <P>
992  <b>int pcre_info(const pcre *<i>code</i>, int *<i>optptr</i>, int</b>  <b>int pcre_info(const pcre *<i>code</i>, int *<i>optptr</i>, int</b>
993  <b>*<i>firstcharptr</i>);</b>  <b>*<i>firstcharptr</i>);</b>
# Line 753  restrictive to return all the available Line 998  restrictive to return all the available
998  programs should use <b>pcre_fullinfo()</b> instead. The yield of  programs should use <b>pcre_fullinfo()</b> instead. The yield of
999  <b>pcre_info()</b> is the number of capturing subpatterns, or one of the  <b>pcre_info()</b> is the number of capturing subpatterns, or one of the
1000  following negative numbers:  following negative numbers:
 </P>  
 <P>  
1001  <pre>  <pre>
1002    PCRE_ERROR_NULL       the argument <i>code</i> was NULL    PCRE_ERROR_NULL       the argument <i>code</i> was NULL
1003    PCRE_ERROR_BADMAGIC   the "magic number" was not found    PCRE_ERROR_BADMAGIC   the "magic number" was not found
1004  </PRE>  </pre>
 </P>  
 <P>  
1005  If the <i>optptr</i> argument is not NULL, a copy of the options with which the  If the <i>optptr</i> argument is not NULL, a copy of the options with which the
1006  pattern was compiled is placed in the integer it points to (see  pattern was compiled is placed in the integer it points to (see
1007  PCRE_INFO_OPTIONS above).  PCRE_INFO_OPTIONS above).
# Line 770  If the pattern is not anchored and the < Line 1011  If the pattern is not anchored and the <
1011  it is used to pass back information about the first character of any matched  it is used to pass back information about the first character of any matched
1012  string (see PCRE_INFO_FIRSTBYTE above).  string (see PCRE_INFO_FIRSTBYTE above).
1013  </P>  </P>
1014  <br><a name="SEC10" href="#TOC1">MATCHING A PATTERN</a><br>  <br><a name="SEC13" href="#TOC1">REFERENCE COUNTS</a><br>
1015    <P>
1016    <b>int pcre_refcount(pcre *<i>code</i>, int <i>adjust</i>);</b>
1017    </P>
1018    <P>
1019    The <b>pcre_refcount()</b> function is used to maintain a reference count in the
1020    data block that contains a compiled pattern. It is provided for the benefit of
1021    applications that operate in an object-oriented manner, where different parts
1022    of the application may be using the same compiled pattern, but you want to free
1023    the block when they are all done.
1024    </P>
1025    <P>
1026    When a pattern is compiled, the reference count field is initialized to zero.
1027    It is changed only by calling this function, whose action is to add the
1028    <i>adjust</i> value (which may be positive or negative) to it. The yield of the
1029    function is the new value. However, the value of the count is constrained to
1030    lie between 0 and 65535, inclusive. If the new value is outside these limits,
1031    it is forced to the appropriate limit value.
1032    </P>
1033    <P>
1034    Except when it is zero, the reference count is not correctly preserved if a
1035    pattern is compiled on one host and then transferred to a host whose byte-order
1036    is different. (This seems a highly unlikely scenario.)
1037    </P>
1038    <br><a name="SEC14" href="#TOC1">MATCHING A PATTERN: THE TRADITIONAL FUNCTION</a><br>
1039  <P>  <P>
1040  <b>int pcre_exec(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>  <b>int pcre_exec(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>
1041  <b>const char *<i>subject</i>, int <i>length</i>, int <i>startoffset</i>,</b>  <b>const char *<i>subject</i>, int <i>length</i>, int <i>startoffset</i>,</b>
# Line 778  string (see PCRE_INFO_FIRSTBYTE above). Line 1043  string (see PCRE_INFO_FIRSTBYTE above).
1043  </P>  </P>
1044  <P>  <P>
1045  The function <b>pcre_exec()</b> is called to match a subject string against a  The function <b>pcre_exec()</b> is called to match a subject string against a
1046  pre-compiled pattern, which is passed in the <i>code</i> argument. If the  compiled pattern, which is passed in the <i>code</i> argument. If the
1047  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
1048  <i>extra</i> argument.  <i>extra</i> argument. This function is the main matching facility of the
1049    library, and it operates in a Perl-like manner. For specialist use there is
1050    also an alternative matching function, which is described
1051    <a href="#dfamatch">below</a>
1052    in the section about the <b>pcre_dfa_exec()</b> function.
1053  </P>  </P>
1054  <P>  <P>
1055  Here is an example of a simple call to <b>pcre_exec()</b>:  In most applications, the pattern will have been compiled (and optionally
1056    studied) in the same process that calls <b>pcre_exec()</b>. However, it is
1057    possible to save compiled patterns and study data, and then use them later
1058    in different processes, possibly even on different hosts. For a discussion
1059    about this, see the
1060    <a href="pcreprecompile.html"><b>pcreprecompile</b></a>
1061    documentation.
1062  </P>  </P>
1063  <P>  <P>
1064    Here is an example of a simple call to <b>pcre_exec()</b>:
1065  <pre>  <pre>
1066    int rc;    int rc;
1067    int ovector[30];    int ovector[30];
# Line 796  Here is an example of a simple call to < Line 1072  Here is an example of a simple call to <
1072      11,             /* the length of the subject string */      11,             /* the length of the subject string */
1073      0,              /* start at offset 0 in the subject */      0,              /* start at offset 0 in the subject */
1074      0,              /* default options */      0,              /* default options */
1075      ovector,        /* vector for substring information */      ovector,        /* vector of integers for substring information */
1076      30);            /* number of elements in the vector */      30);            /* number of elements (NOT size in bytes) */
1077  </PRE>  <a name="extradata"></a></PRE>
1078  </P>  </P>
1079    <br><b>
1080    Extra data for <b>pcre_exec()</b>
1081    </b><br>
1082  <P>  <P>
1083  If the <i>extra</i> argument is not NULL, it must point to a <b>pcre_extra</b>  If the <i>extra</i> argument is not NULL, it must point to a <b>pcre_extra</b>
1084  data block. The <b>pcre_study()</b> function returns such a block (when it  data block. The <b>pcre_study()</b> function returns such a block (when it
1085  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
1086  additional information in it. The fields in the block are as follows:  additional information in it. The <b>pcre_extra</b> block contains the following
1087  </P>  fields (not necessarily in this order):
 <P>  
1088  <pre>  <pre>
1089    unsigned long int <i>flags</i>;    unsigned long int <i>flags</i>;
1090    void *<i>study_data</i>;    void *<i>study_data</i>;
1091    unsigned long int <i>match_limit</i>;    unsigned long int <i>match_limit</i>;
1092      unsigned long int <i>match_limit_recursion</i>;
1093    void *<i>callout_data</i>;    void *<i>callout_data</i>;
1094  </PRE>    const unsigned char *<i>tables</i>;
1095  </P>  </pre>
 <P>  
1096  The <i>flags</i> field is a bitmap that specifies which of the other fields  The <i>flags</i> field is a bitmap that specifies which of the other fields
1097  are set. The flag bits are:  are set. The flag bits are:
 </P>  
 <P>  
1098  <pre>  <pre>
1099    PCRE_EXTRA_STUDY_DATA    PCRE_EXTRA_STUDY_DATA
1100    PCRE_EXTRA_MATCH_LIMIT    PCRE_EXTRA_MATCH_LIMIT
1101      PCRE_EXTRA_MATCH_LIMIT_RECURSION
1102    PCRE_EXTRA_CALLOUT_DATA    PCRE_EXTRA_CALLOUT_DATA
1103  </PRE>    PCRE_EXTRA_TABLES
1104  </P>  </pre>
 <P>  
1105  Other flag bits should be set to zero. The <i>study_data</i> field is set in the  Other flag bits should be set to zero. The <i>study_data</i> field is set in the
1106  <b>pcre_extra</b> block that is returned by <b>pcre_study()</b>, together with  <b>pcre_extra</b> block that is returned by <b>pcre_study()</b>, together with
1107  the appropriate flag bit. You should not set this yourself, but you can add to  the appropriate flag bit. You should not set this yourself, but you may add to
1108  the block by setting the other fields.  the block by setting the other fields and their corresponding flag bits.
1109  </P>  </P>
1110  <P>  <P>
1111  The <i>match_limit</i> field provides a means of preventing PCRE from using up a  The <i>match_limit</i> field provides a means of preventing PCRE from using up a
1112  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,
1113  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
1114  classic example is the use of nested unlimited repeats. Internally, PCRE uses a  classic example is the use of nested unlimited repeats.
 function called <b>match()</b> which it calls repeatedly (sometimes  
 recursively). The limit is imposed on the number of times this function is  
 called during a match, which has the effect of limiting the amount of recursion  
 and backtracking that can take place. For patterns that are not anchored, the  
 count starts from zero for each position in the subject string.  
1115  </P>  </P>
1116  <P>  <P>
1117  The default limit for the library can be set when PCRE is built; the default  Internally, PCRE uses a function called <b>match()</b> which it calls repeatedly
1118  default is 10 million, which handles all but the most extreme cases. You can  (sometimes recursively). The limit set by <i>match_limit</i> is imposed on the
1119  reduce the default by suppling <b>pcre_exec()</b> with a \fRpcre_extra\fR block  number of times this function is called during a match, which has the effect of
1120  in which <i>match_limit</i> is set to a smaller value, and  limiting the amount of backtracking that can take place. For patterns that are
1121  PCRE_EXTRA_MATCH_LIMIT is set in the <i>flags</i> field. If the limit is  not anchored, the count restarts from zero for each position in the subject
1122  exceeded, <b>pcre_exec()</b> returns PCRE_ERROR_MATCHLIMIT.  string.
1123  </P>  </P>
1124  <P>  <P>
1125  The <i>pcre_callout</i> field is used in conjunction with the "callout" feature,  The default value for the limit can be set when PCRE is built; the default
1126  which is described in the <b>pcrecallout</b> documentation.  default is 10 million, which handles all but the most extreme cases. You can
1127  </P>  override the default by suppling <b>pcre_exec()</b> with a <b>pcre_extra</b>
1128  <P>  block in which <i>match_limit</i> is set, and PCRE_EXTRA_MATCH_LIMIT is set in
1129  The PCRE_ANCHORED option can be passed in the <i>options</i> argument, whose  the <i>flags</i> field. If the limit is exceeded, <b>pcre_exec()</b> returns
1130  unused bits must be zero. This limits <b>pcre_exec()</b> to matching at the  PCRE_ERROR_MATCHLIMIT.
1131  first matching position. However, if a pattern was compiled with PCRE_ANCHORED,  </P>
1132  or turned out to be anchored by virtue of its contents, it cannot be made  <P>
1133  unachored at matching time.  The <i>match_limit_recursion</i> field is similar to <i>match_limit</i>, but
1134    instead of limiting the total number of times that <b>match()</b> is called, it
1135    limits the depth of recursion. The recursion depth is a smaller number than the
1136    total number of calls, because not all calls to <b>match()</b> are recursive.
1137    This limit is of use only if it is set smaller than <i>match_limit</i>.
1138    </P>
1139    <P>
1140    Limiting the recursion depth limits the amount of stack that can be used, or,
1141    when PCRE has been compiled to use memory on the heap instead of the stack, the
1142    amount of heap memory that can be used.
1143    </P>
1144    <P>
1145    The default value for <i>match_limit_recursion</i> can be set when PCRE is
1146    built; the default default is the same value as the default for
1147    <i>match_limit</i>. You can override the default by suppling <b>pcre_exec()</b>
1148    with a <b>pcre_extra</b> block in which <i>match_limit_recursion</i> is set, and
1149    PCRE_EXTRA_MATCH_LIMIT_RECURSION is set in the <i>flags</i> field. If the limit
1150    is exceeded, <b>pcre_exec()</b> returns PCRE_ERROR_RECURSIONLIMIT.
1151  </P>  </P>
1152  <P>  <P>
1153  There are also three further options that can be set only at matching time:  The <i>pcre_callout</i> field is used in conjunction with the "callout" feature,
1154    which is described in the
1155    <a href="pcrecallout.html"><b>pcrecallout</b></a>
1156    documentation.
1157  </P>  </P>
1158  <P>  <P>
1159    The <i>tables</i> field is used to pass a character tables pointer to
1160    <b>pcre_exec()</b>; this overrides the value that is stored with the compiled
1161    pattern. A non-NULL value is stored with the compiled pattern only if custom
1162    tables were supplied to <b>pcre_compile()</b> via its <i>tableptr</i> argument.
1163    If NULL is passed to <b>pcre_exec()</b> using this mechanism, it forces PCRE's
1164    internal tables to be used. This facility is helpful when re-using patterns
1165    that have been saved after compiling with an external set of tables, because
1166    the external tables might be at a different address when <b>pcre_exec()</b> is
1167    called. See the
1168    <a href="pcreprecompile.html"><b>pcreprecompile</b></a>
1169    documentation for a discussion of saving compiled patterns for later use.
1170    </P>
1171    <br><b>
1172    Option bits for <b>pcre_exec()</b>
1173    </b><br>
1174    <P>
1175    The unused bits of the <i>options</i> argument for <b>pcre_exec()</b> must be
1176    zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_<i>xxx</i>,
1177    PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NO_UTF8_CHECK and PCRE_PARTIAL.
1178    <pre>
1179      PCRE_ANCHORED
1180    </pre>
1181    The PCRE_ANCHORED option limits <b>pcre_exec()</b> to matching at the first
1182    matching position. If a pattern was compiled with PCRE_ANCHORED, or turned out
1183    to be anchored by virtue of its contents, it cannot be made unachored at
1184    matching time.
1185    <pre>
1186      PCRE_NEWLINE_CR
1187      PCRE_NEWLINE_LF
1188      PCRE_NEWLINE_CRLF
1189      PCRE_NEWLINE_ANYCRLF
1190      PCRE_NEWLINE_ANY
1191    </pre>
1192    These options override the newline definition that was chosen or defaulted when
1193    the pattern was compiled. For details, see the description of
1194    <b>pcre_compile()</b> above. During matching, the newline choice affects the
1195    behaviour of the dot, circumflex, and dollar metacharacters. It may also alter
1196    the way the match position is advanced after a match failure for an unanchored
1197    pattern. When PCRE_NEWLINE_CRLF, PCRE_NEWLINE_ANYCRLF, or PCRE_NEWLINE_ANY is
1198    set, and a match attempt fails when the current position is at a CRLF sequence,
1199    the match position is advanced by two characters instead of one, in other
1200    words, to after the CRLF.
1201  <pre>  <pre>
1202    PCRE_NOTBOL    PCRE_NOTBOL
1203  </PRE>  </pre>
1204  </P>  This option specifies that first character of the subject string is not the
1205  <P>  beginning of a line, so the circumflex metacharacter should not match before
1206  The first character of the string is not the beginning of a line, so the  it. Setting this without PCRE_MULTILINE (at compile time) causes circumflex
1207  circumflex metacharacter should not match before it. Setting this without  never to match. This option affects only the behaviour of the circumflex
1208  PCRE_MULTILINE (at compile time) causes circumflex never to match.  metacharacter. It does not affect \A.
 </P>  
 <P>  
1209  <pre>  <pre>
1210    PCRE_NOTEOL    PCRE_NOTEOL
1211  </PRE>  </pre>
1212  </P>  This option specifies that the end of the subject string is not the end of a
1213  <P>  line, so the dollar metacharacter should not match it nor (except in multiline
1214  The end of the string is not the end of a line, so the dollar metacharacter  mode) a newline immediately before it. Setting this without PCRE_MULTILINE (at
1215  should not match it nor (except in multiline mode) a newline immediately before  compile time) causes dollar never to match. This option affects only the
1216  it. Setting this without PCRE_MULTILINE (at compile time) causes dollar never  behaviour of the dollar metacharacter. It does not affect \Z or \z.
 to match.  
 </P>  
 <P>  
1217  <pre>  <pre>
1218    PCRE_NOTEMPTY    PCRE_NOTEMPTY
1219  </PRE>  </pre>
 </P>  
 <P>  
1220  An empty string is not considered to be a valid match if this option is set. If  An empty string is not considered to be a valid match if this option is set. If
1221  there are alternatives in the pattern, they are tried. If all the alternatives  there are alternatives in the pattern, they are tried. If all the alternatives
1222  match the empty string, the entire match fails. For example, if the pattern  match the empty string, the entire match fails. For example, if the pattern
 </P>  
 <P>  
1223  <pre>  <pre>
1224    a?b?    a?b?
1225  </PRE>  </pre>
 </P>  
 <P>  
1226  is applied to a string not beginning with "a" or "b", it matches the empty  is applied to a string not beginning with "a" or "b", it matches the empty
1227  string at the start of the subject. With PCRE_NOTEMPTY set, this match is not  string at the start of the subject. With PCRE_NOTEMPTY set, this match is not
1228  valid, so PCRE searches further into the string for occurrences of "a" or "b".  valid, so PCRE searches further into the string for occurrences of "a" or "b".
# Line 910  Perl has no direct equivalent of PCRE_NO Line 1232  Perl has no direct equivalent of PCRE_NO
1232  of a pattern match of the empty string within its <b>split()</b> function, and  of a pattern match of the empty string within its <b>split()</b> function, and
1233  when using the /g modifier. It is possible to emulate Perl's behaviour after  when using the /g modifier. It is possible to emulate Perl's behaviour after
1234  matching a null string by first trying the match again at the same offset with  matching a null string by first trying the match again at the same offset with
1235  PCRE_NOTEMPTY set, and then if that fails by advancing the starting offset (see  PCRE_NOTEMPTY and PCRE_ANCHORED, and then if that fails by advancing the
1236  below) and trying an ordinary match again.  starting offset (see below) and trying an ordinary match again. There is some
1237    code that demonstrates how to do this in the <i>pcredemo.c</i> sample program.
1238    <pre>
1239      PCRE_NO_UTF8_CHECK
1240    </pre>
1241    When PCRE_UTF8 is set at compile time, the validity of the subject as a UTF-8
1242    string is automatically checked when <b>pcre_exec()</b> is subsequently called.
1243    The value of <i>startoffset</i> is also checked to ensure that it points to the
1244    start of a UTF-8 character. If an invalid UTF-8 sequence of bytes is found,
1245    <b>pcre_exec()</b> returns the error PCRE_ERROR_BADUTF8. If <i>startoffset</i>
1246    contains an invalid value, PCRE_ERROR_BADUTF8_OFFSET is returned.
1247    </P>
1248    <P>
1249    If you already know that your subject is valid, and you want to skip these
1250    checks for performance reasons, you can set the PCRE_NO_UTF8_CHECK option when
1251    calling <b>pcre_exec()</b>. You might want to do this for the second and
1252    subsequent calls to <b>pcre_exec()</b> if you are making repeated calls to find
1253    all the matches in a single subject string. However, you should be sure that
1254    the value of <i>startoffset</i> points to the start of a UTF-8 character. When
1255    PCRE_NO_UTF8_CHECK is set, the effect of passing an invalid UTF-8 string as a
1256    subject, or a value of <i>startoffset</i> that does not point to the start of a
1257    UTF-8 character, is undefined. Your program may crash.
1258    <pre>
1259      PCRE_PARTIAL
1260    </pre>
1261    This option turns on the partial matching feature. If the subject string fails
1262    to match the pattern, but at some point during the matching process the end of
1263    the subject was reached (that is, the subject partially matches the pattern and
1264    the failure to match occurred only because there were not enough subject
1265    characters), <b>pcre_exec()</b> returns PCRE_ERROR_PARTIAL instead of
1266    PCRE_ERROR_NOMATCH. When PCRE_PARTIAL is used, there are restrictions on what
1267    may appear in the pattern. These are discussed in the
1268    <a href="pcrepartial.html"><b>pcrepartial</b></a>
1269    documentation.
1270  </P>  </P>
1271    <br><b>
1272    The string to be matched by <b>pcre_exec()</b>
1273    </b><br>
1274  <P>  <P>
1275  The subject string is passed to <b>pcre_exec()</b> as a pointer in  The subject string is passed to <b>pcre_exec()</b> as a pointer in
1276  <i>subject</i>, a length in <i>length</i>, and a starting offset in  <i>subject</i>, a length in <i>length</i>, and a starting byte offset in
1277  <i>startoffset</i>. Unlike the pattern string, the subject may contain binary  <i>startoffset</i>. In UTF-8 mode, the byte offset must point to the start of a
1278  zero bytes. When the starting offset is zero, the search for a match starts at  UTF-8 character. Unlike the pattern string, the subject may contain binary zero
1279  the beginning of the subject, and this is by far the most common case.  bytes. When the starting offset is zero, the search for a match starts at the
1280  </P>  beginning of the subject, and this is by far the most common case.
 <P>  
 If the pattern was compiled with the PCRE_UTF8 option, the subject must be a  
 sequence of bytes that is a valid UTF-8 string. If an invalid UTF-8 string is  
 passed, PCRE's behaviour is not defined.  
1281  </P>  </P>
1282  <P>  <P>
1283  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
# Line 931  same subject by calling <b>pcre_exec()</ Line 1285  same subject by calling <b>pcre_exec()</
1285  Setting <i>startoffset</i> differs from just passing over a shortened string and  Setting <i>startoffset</i> differs from just passing over a shortened string and
1286  setting PCRE_NOTBOL in the case of a pattern that begins with any kind of  setting PCRE_NOTBOL in the case of a pattern that begins with any kind of
1287  lookbehind. For example, consider the pattern  lookbehind. For example, consider the pattern
 </P>  
 <P>  
1288  <pre>  <pre>
1289    \Biss\B    \Biss\B
1290  </PRE>  </pre>
 </P>  
 <P>  
1291  which finds occurrences of "iss" in the middle of words. (\B matches only if  which finds occurrences of "iss" in the middle of words. (\B matches only if
1292  the current position in the subject is not a word boundary.) When applied to  the current position in the subject is not a word boundary.) When applied to
1293  the string "Mississipi" the first call to <b>pcre_exec()</b> finds the first  the string "Mississipi" the first call to <b>pcre_exec()</b> finds the first
# Line 950  behind the starting point to discover th Line 1300  behind the starting point to discover th
1300  </P>  </P>
1301  <P>  <P>
1302  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
1303  attempt to match at the given offset is tried. This can only succeed if the  attempt to match at the given offset is made. This can only succeed if the
1304  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.
1305  </P>  </P>
1306    <br><b>
1307    How <b>pcre_exec()</b> returns captured substrings
1308    </b><br>
1309  <P>  <P>
1310  In general, a pattern matches a certain portion of the subject, and in  In general, a pattern matches a certain portion of the subject, and in
1311  addition, further substrings from the subject may be picked out by parts of the  addition, further substrings from the subject may be picked out by parts of the
# Line 964  kinds of parenthesized subpattern that d Line 1317  kinds of parenthesized subpattern that d
1317  <P>  <P>
1318  Captured substrings are returned to the caller via a vector of integer offsets  Captured substrings are returned to the caller via a vector of integer offsets
1319  whose address is passed in <i>ovector</i>. The number of elements in the vector  whose address is passed in <i>ovector</i>. The number of elements in the vector
1320  is passed in <i>ovecsize</i>. The first two-thirds of the vector is used to pass  is passed in <i>ovecsize</i>, which must be a non-negative number. <b>Note</b>:
1321  back captured substrings, each substring using a pair of integers. The  this argument is NOT the size of <i>ovector</i> in bytes.
1322  remaining third of the vector is used as workspace by <b>pcre_exec()</b> while  </P>
1323  matching capturing subpatterns, and is not available for passing back  <P>
1324  information. The length passed in <i>ovecsize</i> should always be a multiple of  The first two-thirds of the vector is used to pass back captured substrings,
1325  three. If it is not, it is rounded down.  each substring using a pair of integers. The remaining third of the vector is
1326    used as workspace by <b>pcre_exec()</b> while matching capturing subpatterns,
1327    and is not available for passing back information. The length passed in
1328    <i>ovecsize</i> should always be a multiple of three. If it is not, it is
1329    rounded down.
1330  </P>  </P>
1331  <P>  <P>
1332  When a match has been successful, information about captured substrings is  When a match is successful, information about captured substrings is returned
1333  returned in pairs of integers, starting at the beginning of <i>ovector</i>, and  in pairs of integers, starting at the beginning of <i>ovector</i>, and
1334  continuing up to two-thirds of its length at the most. The first element of a  continuing up to two-thirds of its length at the most. The first element of a
1335  pair is set to the offset of the first character in a substring, and the second  pair is set to the offset of the first character in a substring, and the second
1336  is set to the offset of the first character after the end of a substring. The  is set to the offset of the first character after the end of a substring. The
1337  first pair, <i>ovector[0]</i> and <i>ovector[1]</i>, identify the portion of the  first pair, <i>ovector[0]</i> and <i>ovector[1]</i>, identify the portion of the
1338  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
1339  first capturing subpattern, and so on. The value returned by <b>pcre_exec()</b>  first capturing subpattern, and so on. The value returned by <b>pcre_exec()</b>
1340  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
1341  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
1342  just the first pair of offsets has been set.  capturing subpatterns, the return value from a successful match is 1,
1343  </P>  indicating that just the first pair of offsets has been set.
 <P>  
 Some convenience functions are provided for extracting the captured substrings  
 as separate strings. These are described in the following section.  
 </P>  
 <P>  
 It is possible for an capturing subpattern number <i>n+1</i> to match some  
 part of the subject when subpattern <i>n</i> 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.  
1344  </P>  </P>
1345  <P>  <P>
1346  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
1347  string that it matched that gets returned.  string that it matched that is returned.
1348  </P>  </P>
1349  <P>  <P>
1350  If the vector is too small to hold all the captured substrings, it is used as  If the vector is too small to hold all the captured substring offsets, it is
1351  far as possible (up to two-thirds of its length), and the function returns a  used as far as possible (up to two-thirds of its length), and the function
1352  value of zero. In particular, if the substring offsets are not of interest,  returns a value of zero. In particular, if the substring offsets are not of
1353  <b>pcre_exec()</b> may be called with <i>ovector</i> passed as NULL and  interest, <b>pcre_exec()</b> may be called with <i>ovector</i> passed as NULL and
1354  <i>ovecsize</i> as zero. However, if the pattern contains back references and  <i>ovecsize</i> as zero. However, if the pattern contains back references and
1355  the <i>ovector</i> isn't big enough to remember the related substrings, PCRE has  the <i>ovector</i> is not big enough to remember the related substrings, PCRE
1356  to get additional memory for use during matching. Thus it is usually advisable  has to get additional memory for use during matching. Thus it is usually
1357  to supply an <i>ovector</i>.  advisable to supply an <i>ovector</i>.
1358  </P>  </P>
1359  <P>  <P>
1360  Note that <b>pcre_info()</b> can be used to find out how many capturing  The <b>pcre_info()</b> function can be used to find out how many capturing
1361  subpatterns there are in a compiled pattern. The smallest size for  subpatterns there are in a compiled pattern. The smallest size for
1362  <i>ovector</i> that will allow for <i>n</i> captured substrings, in addition to  <i>ovector</i> that will allow for <i>n</i> captured substrings, in addition to
1363  the offsets of the substring matched by the whole pattern, is (<i>n</i>+1)*3.  the offsets of the substring matched by the whole pattern, is (<i>n</i>+1)*3.
1364  </P>  </P>
1365  <P>  <P>
1366  If <b>pcre_exec()</b> fails, it returns a negative number. The following are  It is possible for capturing subpattern number <i>n+1</i> to match some part of
1367  defined in the header file:  the subject when subpattern <i>n</i> has not been used at all. For example, if
1368    the string "abc" is matched against the pattern (a|(z))(bc) the return from the
1369    function is 4, and subpatterns 1 and 3 are matched, but 2 is not. When this
1370    happens, both values in the offset pairs corresponding to unused subpatterns
1371    are set to -1.
1372  </P>  </P>
1373  <P>  <P>
1374  <pre>  Offset values that correspond to unused subpatterns at the end of the
1375    PCRE_ERROR_NOMATCH        (-1)  expression are also set to -1. For example, if the string "abc" is matched
1376  </PRE>  against the pattern (abc)(x(yz)?)? subpatterns 2 and 3 are not matched. The
1377    return from the function is 2, because the highest used capturing subpattern
1378    number is 1. However, you can refer to the offsets for the second and third
1379    capturing subpatterns if you wish (assuming the vector is large enough, of
1380    course).
1381  </P>  </P>
1382  <P>  <P>
1383  The subject string did not match the pattern.  Some convenience functions are provided for extracting the captured substrings
1384  </P>  as separate strings. These are described below.
1385    <a name="errorlist"></a></P>
1386    <br><b>
1387    Error return values from <b>pcre_exec()</b>
1388    </b><br>
1389  <P>  <P>
1390    If <b>pcre_exec()</b> fails, it returns a negative number. The following are
1391    defined in the header file:
1392    <pre>
1393      PCRE_ERROR_NOMATCH        (-1)
1394    </pre>
1395    The subject string did not match the pattern.
1396  <pre>  <pre>
1397    PCRE_ERROR_NULL           (-2)    PCRE_ERROR_NULL           (-2)
1398  </PRE>  </pre>
 </P>  
 <P>  
1399  Either <i>code</i> or <i>subject</i> was passed as NULL, or <i>ovector</i> was  Either <i>code</i> or <i>subject</i> was passed as NULL, or <i>ovector</i> was
1400  NULL and <i>ovecsize</i> was not zero.  NULL and <i>ovecsize</i> was not zero.
 </P>  
 <P>  
1401  <pre>  <pre>
1402    PCRE_ERROR_BADOPTION      (-3)    PCRE_ERROR_BADOPTION      (-3)
1403  </PRE>  </pre>
 </P>  
 <P>  
1404  An unrecognized bit was set in the <i>options</i> argument.  An unrecognized bit was set in the <i>options</i> argument.
 </P>  
 <P>  
1405  <pre>  <pre>
1406    PCRE_ERROR_BADMAGIC       (-4)    PCRE_ERROR_BADMAGIC       (-4)
1407  </PRE>  </pre>
 </P>  
 <P>  
1408  PCRE stores a 4-byte "magic number" at the start of the compiled code, to catch  PCRE stores a 4-byte "magic number" at the start of the compiled code, to catch
1409  the case when it is passed a junk pointer. This is the error it gives when the  the case when it is passed a junk pointer and to detect when a pattern that was
1410  magic number isn't present.  compiled in an environment of one endianness is run in an environment with the
1411  </P>  other endianness. This is the error that PCRE gives when the magic number is
1412  <P>  not present.
1413  <pre>  <pre>
1414    PCRE_ERROR_UNKNOWN_NODE   (-5)    PCRE_ERROR_UNKNOWN_OPCODE (-5)
1415  </PRE>  </pre>
 </P>  
 <P>  
1416  While running the pattern match, an unknown item was encountered in the  While running the pattern match, an unknown item was encountered in the
1417  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
1418  of the compiled pattern.  of the compiled pattern.
 </P>  
 <P>  
1419  <pre>  <pre>
1420    PCRE_ERROR_NOMEMORY       (-6)    PCRE_ERROR_NOMEMORY       (-6)
1421  </PRE>  </pre>
 </P>  
 <P>  
1422  If a pattern contains back references, but the <i>ovector</i> that is passed to  If a pattern contains back references, but the <i>ovector</i> that is passed to
1423  <b>pcre_exec()</b> is not big enough to remember the referenced substrings, PCRE  <b>pcre_exec()</b> is not big enough to remember the referenced substrings, PCRE
1424  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
1425  call via <b>pcre_malloc()</b> fails, this error is given. The memory is freed at  call via <b>pcre_malloc()</b> fails, this error is given. The memory is
1426  the end of matching.  automatically freed at the end of matching.
 </P>  
 <P>  
1427  <pre>  <pre>
1428    PCRE_ERROR_NOSUBSTRING    (-7)    PCRE_ERROR_NOSUBSTRING    (-7)
1429  </PRE>  </pre>
 </P>  
 <P>  
1430  This error is used by the <b>pcre_copy_substring()</b>,  This error is used by the <b>pcre_copy_substring()</b>,
1431  <b>pcre_get_substring()</b>, and <b>pcre_get_substring_list()</b> functions (see  <b>pcre_get_substring()</b>, and <b>pcre_get_substring_list()</b> functions (see
1432  below). It is never returned by <b>pcre_exec()</b>.  below). It is never returned by <b>pcre_exec()</b>.
 </P>  
 <P>  
1433  <pre>  <pre>
1434    PCRE_ERROR_MATCHLIMIT     (-8)    PCRE_ERROR_MATCHLIMIT     (-8)
1435  </PRE>  </pre>
1436  </P>  The backtracking limit, as specified by the <i>match_limit</i> field in a
1437  <P>  <b>pcre_extra</b> structure (or defaulted) was reached. See the description
1438  The recursion and backtracking limit, as specified by the <i>match_limit</i>  above.
1439    <pre>
1440      PCRE_ERROR_CALLOUT        (-9)
1441    </pre>
1442    This error is never generated by <b>pcre_exec()</b> itself. It is provided for
1443    use by callout functions that want to yield a distinctive error code. See the
1444    <a href="pcrecallout.html"><b>pcrecallout</b></a>
1445    documentation for details.
1446    <pre>
1447      PCRE_ERROR_BADUTF8        (-10)
1448    </pre>
1449    A string that contains an invalid UTF-8 byte sequence was passed as a subject.
1450    <pre>
1451      PCRE_ERROR_BADUTF8_OFFSET (-11)
1452    </pre>
1453    The UTF-8 byte sequence that was passed as a subject was valid, but the value
1454    of <i>startoffset</i> did not point to the beginning of a UTF-8 character.
1455    <pre>
1456      PCRE_ERROR_PARTIAL        (-12)
1457    </pre>
1458    The subject string did not match, but it did match partially. See the
1459    <a href="pcrepartial.html"><b>pcrepartial</b></a>
1460    documentation for details of partial matching.
1461    <pre>
1462      PCRE_ERROR_BADPARTIAL     (-13)
1463    </pre>
1464    The PCRE_PARTIAL option was used with a compiled pattern containing items that
1465    are not supported for partial matching. See the
1466    <a href="pcrepartial.html"><b>pcrepartial</b></a>
1467    documentation for details of partial matching.
1468    <pre>
1469      PCRE_ERROR_INTERNAL       (-14)
1470    </pre>
1471    An unexpected internal error has occurred. This error could be caused by a bug
1472    in PCRE or by overwriting of the compiled pattern.
1473    <pre>
1474      PCRE_ERROR_BADCOUNT       (-15)
1475    </pre>
1476    This error is given if the value of the <i>ovecsize</i> argument is negative.
1477    <pre>
1478      PCRE_ERROR_RECURSIONLIMIT (-21)
1479    </pre>
1480    The internal recursion limit, as specified by the <i>match_limit_recursion</i>
1481  field in a <b>pcre_extra</b> structure (or defaulted) was reached. See the  field in a <b>pcre_extra</b> structure (or defaulted) was reached. See the
1482  description above.  description above.
 </P>  
 <P>  
1483  <pre>  <pre>
1484    PCRE_ERROR_CALLOUT        (-9)    PCRE_ERROR_NULLWSLIMIT    (-22)
1485  </PRE>  </pre>
1486    When a group that can match an empty substring is repeated with an unbounded
1487    upper limit, the subject position at the start of the group must be remembered,
1488    so that a test for an empty string can be made when the end of the group is
1489    reached. Some workspace is required for this; if it runs out, this error is
1490    given.
1491    <pre>
1492      PCRE_ERROR_BADNEWLINE     (-23)
1493    </pre>
1494    An invalid combination of PCRE_NEWLINE_<i>xxx</i> options was given.
1495  </P>  </P>
1496  <P>  <P>
1497  This error is never generated by <b>pcre_exec()</b> itself. It is provided for  Error numbers -16 to -20 are not used by <b>pcre_exec()</b>.
 use by callout functions that want to yield a distinctive error code. See the  
 <b>pcrecallout</b> documentation for details.  
1498  </P>  </P>
1499  <br><a name="SEC11" href="#TOC1">EXTRACTING CAPTURED SUBSTRINGS BY NUMBER</a><br>  <br><a name="SEC15" href="#TOC1">EXTRACTING CAPTURED SUBSTRINGS BY NUMBER</a><br>
1500  <P>  <P>
1501  <b>int pcre_copy_substring(const char *<i>subject</i>, int *<i>ovector</i>,</b>  <b>int pcre_copy_substring(const char *<i>subject</i>, int *<i>ovector</i>,</b>
1502  <b>int <i>stringcount</i>, int <i>stringnumber</i>, char *<i>buffer</i>,</b>  <b>int <i>stringcount</i>, int <i>stringnumber</i>, char *<i>buffer</i>,</b>
# Line 1128  Captured substrings can be accessed dire Line 1518  Captured substrings can be accessed dire
1518  <b>pcre_get_substring_list()</b> are provided for extracting captured substrings  <b>pcre_get_substring_list()</b> are provided for extracting captured substrings
1519  as new, separate, zero-terminated strings. These functions identify substrings  as new, separate, zero-terminated strings. These functions identify substrings
1520  by number. The next section describes functions for extracting named  by number. The next section describes functions for extracting named
1521  substrings. A substring that contains a binary zero is correctly extracted and  substrings.
1522  has a further zero added on the end, but the result is not, of course,  </P>
1523  a C string.  <P>
1524    A substring that contains a binary zero is correctly extracted and has a
1525    further zero added on the end, but the result is not, of course, a C string.
1526    However, you can process such a string by referring to the length that is
1527    returned by <b>pcre_copy_substring()</b> and <b>pcre_get_substring()</b>.
1528    Unfortunately, the interface to <b>pcre_get_substring_list()</b> is not adequate
1529    for handling strings containing binary zeros, because the end of the final
1530    string is not independently indicated.
1531  </P>  </P>
1532  <P>  <P>
1533  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:
1534  <i>subject</i> is the subject string which has just been successfully matched,  <i>subject</i> is the subject string that has just been successfully matched,
1535  <i>ovector</i> is a pointer to the vector of integer offsets that was passed to  <i>ovector</i> is a pointer to the vector of integer offsets that was passed to
1536  <b>pcre_exec()</b>, and <i>stringcount</i> is the number of substrings that were  <b>pcre_exec()</b>, and <i>stringcount</i> is the number of substrings that were
1537  captured by the match, including the substring that matched the entire regular  captured by the match, including the substring that matched the entire regular
1538  expression. This is the value returned by <b>pcre_exec</b> if it is greater than  expression. This is the value returned by <b>pcre_exec()</b> if it is greater
1539  zero. If <b>pcre_exec()</b> returned zero, indicating that it ran out of space  than zero. If <b>pcre_exec()</b> returned zero, indicating that it ran out of
1540  in <i>ovector</i>, the value passed as <i>stringcount</i> should be the size of  space in <i>ovector</i>, the value passed as <i>stringcount</i> should be the
1541  the vector divided by three.  number of elements in the vector divided by three.
1542  </P>  </P>
1543  <P>  <P>
1544  The functions <b>pcre_copy_substring()</b> and <b>pcre_get_substring()</b>  The functions <b>pcre_copy_substring()</b> and <b>pcre_get_substring()</b>
1545  extract a single substring, whose number is given as <i>stringnumber</i>. A  extract a single substring, whose number is given as <i>stringnumber</i>. A
1546  value of zero extracts the substring that matched the entire pattern, while  value of zero extracts the substring that matched the entire pattern, whereas
1547  higher values extract the captured substrings. For <b>pcre_copy_substring()</b>,  higher values extract the captured substrings. For <b>pcre_copy_substring()</b>,
1548  the string is placed in <i>buffer</i>, whose length is given by  the string is placed in <i>buffer</i>, whose length is given by
1549  <i>buffersize</i>, while for <b>pcre_get_substring()</b> a new block of memory is  <i>buffersize</i>, while for <b>pcre_get_substring()</b> a new block of memory is
1550  obtained via <b>pcre_malloc</b>, and its address is returned via  obtained via <b>pcre_malloc</b>, and its address is returned via
1551  <i>stringptr</i>. The yield of the function is the length of the string, not  <i>stringptr</i>. The yield of the function is the length of the string, not
1552  including the terminating zero, or one of  including the terminating zero, or one of these error codes:
 </P>  
 <P>  
1553  <pre>  <pre>
1554    PCRE_ERROR_NOMEMORY       (-6)    PCRE_ERROR_NOMEMORY       (-6)
1555  </PRE>  </pre>
 </P>  
 <P>  
1556  The buffer was too small for <b>pcre_copy_substring()</b>, or the attempt to get  The buffer was too small for <b>pcre_copy_substring()</b>, or the attempt to get
1557  memory failed for <b>pcre_get_substring()</b>.  memory failed for <b>pcre_get_substring()</b>.
 </P>  
 <P>  
1558  <pre>  <pre>
1559    PCRE_ERROR_NOSUBSTRING    (-7)    PCRE_ERROR_NOSUBSTRING    (-7)
1560  </PRE>  </pre>
 </P>  
 <P>  
1561  There is no substring whose number is <i>stringnumber</i>.  There is no substring whose number is <i>stringnumber</i>.
1562  </P>  </P>
1563  <P>  <P>
1564  The <b>pcre_get_substring_list()</b> function extracts all available substrings  The <b>pcre_get_substring_list()</b> function extracts all available substrings
1565  and builds a list of pointers to them. All this is done in a single block of  and builds a list of pointers to them. All this is done in a single block of
1566  memory which is obtained via <b>pcre_malloc</b>. The address of the memory block  memory that is obtained via <b>pcre_malloc</b>. The address of the memory block
1567  is returned via <i>listptr</i>, which is also the start of the list of string  is returned via <i>listptr</i>, which is also the start of the list of string
1568  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
1569  function is zero if all went well, or  function is zero if all went well, or the error code
 </P>  
 <P>  
1570  <pre>  <pre>
1571    PCRE_ERROR_NOMEMORY       (-6)    PCRE_ERROR_NOMEMORY       (-6)
1572  </PRE>  </pre>
 </P>  
 <P>  
1573  if the attempt to get the memory block failed.  if the attempt to get the memory block failed.
1574  </P>  </P>
1575  <P>  <P>
# Line 1202  a previous call of <b>pcre_get_substring Line 1587  a previous call of <b>pcre_get_substring
1587  <b>pcre_get_substring_list()</b>, respectively. They do nothing more than call  <b>pcre_get_substring_list()</b>, respectively. They do nothing more than call
1588  the function pointed to by <b>pcre_free</b>, which of course could be called  the function pointed to by <b>pcre_free</b>, which of course could be called
1589  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
1590  linked via a special interface to another programming language which cannot use  linked via a special interface to another programming language that cannot use
1591  <b>pcre_free</b> directly; it is for these cases that the functions are  <b>pcre_free</b> directly; it is for these cases that the functions are
1592  provided.  provided.
1593  </P>  </P>
1594  <br><a name="SEC12" href="#TOC1">EXTRACTING CAPTURED SUBSTRINGS BY NAME</a><br>  <br><a name="SEC16" href="#TOC1">EXTRACTING CAPTURED SUBSTRINGS BY NAME</a><br>
1595    <P>
1596    <b>int pcre_get_stringnumber(const pcre *<i>code</i>,</b>
1597    <b>const char *<i>name</i>);</b>
1598    </P>
1599  <P>  <P>
1600  <b>int pcre_copy_named_substring(const pcre *<i>code</i>,</b>  <b>int pcre_copy_named_substring(const pcre *<i>code</i>,</b>
1601  <b>const char *<i>subject</i>, int *<i>ovector</i>,</b>  <b>const char *<i>subject</i>, int *<i>ovector</i>,</b>
# Line 1214  provided. Line 1603  provided.
1603  <b>char *<i>buffer</i>, int <i>buffersize</i>);</b>  <b>char *<i>buffer</i>, int <i>buffersize</i>);</b>
1604  </P>  </P>
1605  <P>  <P>
 <b>int pcre_get_stringnumber(const pcre *<i>code</i>,</b>  
 <b>const char *<i>name</i>);</b>  
 </P>  
 <P>  
1606  <b>int pcre_get_named_substring(const pcre *<i>code</i>,</b>  <b>int pcre_get_named_substring(const pcre *<i>code</i>,</b>
1607  <b>const char *<i>subject</i>, int *<i>ovector</i>,</b>  <b>const char *<i>subject</i>, int *<i>ovector</i>,</b>
1608  <b>int <i>stringcount</i>, const char *<i>stringname</i>,</b>  <b>int <i>stringcount</i>, const char *<i>stringname</i>,</b>
1609  <b>const char **<i>stringptr</i>);</b>  <b>const char **<i>stringptr</i>);</b>
1610  </P>  </P>
1611  <P>  <P>
1612  To extract a substring by name, you first have to find associated number. This  To extract a substring by name, you first have to find associated number.
1613  can be done by calling <b>pcre_get_stringnumber()</b>. The first argument is the  For example, for this pattern
 compiled pattern, and the second is the name. For example, for this pattern  
 </P>  
 <P>  
1614  <pre>  <pre>
1615    ab(?&#60;xxx&#62;\d+)...    (a+)b(?&#60;xxx&#62;\d+)...
1616  </PRE>  </pre>
1617  </P>  the number of the subpattern called "xxx" is 2. If the name is known to be
1618  <P>  unique (PCRE_DUPNAMES was not set), you can find the number from the name by
1619  the number of the subpattern called "xxx" is 1. Given the number, you can then  calling <b>pcre_get_stringnumber()</b>. The first argument is the compiled
1620  extract the substring directly, or use one of the functions described in the  pattern, and the second is the name. The yield of the function is the
1621  previous section. For convenience, there are also two functions that do the  subpattern number, or PCRE_ERROR_NOSUBSTRING (-7) if there is no subpattern of
1622  whole job.  that name.
1623  </P>  </P>
1624  <P>  <P>
1625  Most of the arguments of <i>pcre_copy_named_substring()</i> and  Given the number, you can extract the substring directly, or use one of the
1626  <i>pcre_get_named_substring()</i> are the same as those for the functions that  functions described in the previous section. For convenience, there are also
1627  extract by number, and so are not re-described here. There are just two  two functions that do the whole job.
1628  differences.  </P>
1629    <P>
1630    Most of the arguments of <b>pcre_copy_named_substring()</b> and
1631    <b>pcre_get_named_substring()</b> are the same as those for the similarly named
1632    functions that extract by number. As these are described in the previous
1633    section, they are not re-described here. There are just two differences:
1634  </P>  </P>
1635  <P>  <P>
1636  First, instead of a substring number, a substring name is given. Second, there  First, instead of a substring number, a substring name is given. Second, there
# Line 1253  translation table. Line 1640  translation table.
1640  </P>  </P>
1641  <P>  <P>
1642  These functions call <b>pcre_get_stringnumber()</b>, and if it succeeds, they  These functions call <b>pcre_get_stringnumber()</b>, and if it succeeds, they
1643  then call <i>pcre_copy_substring()</i> or <i>pcre_get_substring()</i>, as  then call <b>pcre_copy_substring()</b> or <b>pcre_get_substring()</b>, as
1644  appropriate.  appropriate. <b>NOTE:</b> If PCRE_DUPNAMES is set and there are duplicate names,
1645    the behaviour may not be what you want (see the next section).
1646    </P>
1647    <br><a name="SEC17" href="#TOC1">DUPLICATE SUBPATTERN NAMES</a><br>
1648    <P>
1649    <b>int pcre_get_stringtable_entries(const pcre *<i>code</i>,</b>
1650    <b>const char *<i>name</i>, char **<i>first</i>, char **<i>last</i>);</b>
1651    </P>
1652    <P>
1653    When a pattern is compiled with the PCRE_DUPNAMES option, names for subpatterns
1654    are not required to be unique. Normally, patterns with duplicate names are such
1655    that in any one match, only one of the named subpatterns participates. An
1656    example is shown in the
1657    <a href="pcrepattern.html"><b>pcrepattern</b></a>
1658    documentation. When duplicates are present, <b>pcre_copy_named_substring()</b>
1659    and <b>pcre_get_named_substring()</b> return the first substring corresponding
1660    to the given name that is set. If none are set, an empty string is returned.
1661    The <b>pcre_get_stringnumber()</b> function returns one of the numbers that are
1662    associated with the name, but it is not defined which it is.
1663    <br>
1664    <br>
1665    If you want to get full details of all captured substrings for a given name,
1666    you must use the <b>pcre_get_stringtable_entries()</b> function. The first
1667    argument is the compiled pattern, and the second is the name. The third and
1668    fourth are pointers to variables which are updated by the function. After it
1669    has run, they point to the first and last entries in the name-to-number table
1670    for the given name. The function itself returns the length of each entry, or
1671    PCRE_ERROR_NOSUBSTRING (-7) if there are none. The format of the table is
1672    described above in the section entitled <i>Information about a pattern</i>.
1673    Given all the relevant entries for the name, you can extract each of their
1674    numbers, and hence the captured data, if any.
1675    </P>
1676    <br><a name="SEC18" href="#TOC1">FINDING ALL POSSIBLE MATCHES</a><br>
1677    <P>
1678    The traditional matching function uses a similar algorithm to Perl, which stops
1679    when it finds the first match, starting at a given point in the subject. If you
1680    want to find all possible matches, or the longest possible match, consider
1681    using the alternative matching function (see below) instead. If you cannot use
1682    the alternative function, but still need to find all possible matches, you
1683    can kludge it up by making use of the callout facility, which is described in
1684    the
1685    <a href="pcrecallout.html"><b>pcrecallout</b></a>
1686    documentation.
1687    </P>
1688    <P>
1689    What you have to do is to insert a callout right at the end of the pattern.
1690    When your callout function is called, extract and save the current matched
1691    substring. Then return 1, which forces <b>pcre_exec()</b> to backtrack and try
1692    other alternatives. Ultimately, when it runs out of matches, <b>pcre_exec()</b>
1693    will yield PCRE_ERROR_NOMATCH.
1694    <a name="dfamatch"></a></P>
1695    <br><a name="SEC19" href="#TOC1">MATCHING A PATTERN: THE ALTERNATIVE FUNCTION</a><br>
1696    <P>
1697    <b>int pcre_dfa_exec(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>
1698    <b>const char *<i>subject</i>, int <i>length</i>, int <i>startoffset</i>,</b>
1699    <b>int <i>options</i>, int *<i>ovector</i>, int <i>ovecsize</i>,</b>
1700    <b>int *<i>workspace</i>, int <i>wscount</i>);</b>
1701    </P>
1702    <P>
1703    The function <b>pcre_dfa_exec()</b> is called to match a subject string against
1704    a compiled pattern, using a matching algorithm that scans the subject string
1705    just once, and does not backtrack. This has different characteristics to the
1706    normal algorithm, and is not compatible with Perl. Some of the features of PCRE
1707    patterns are not supported. Nevertheless, there are times when this kind of
1708    matching can be useful. For a discussion of the two matching algorithms, see
1709    the
1710    <a href="pcrematching.html"><b>pcrematching</b></a>
1711    documentation.
1712    </P>
1713    <P>
1714    The arguments for the <b>pcre_dfa_exec()</b> function are the same as for
1715    <b>pcre_exec()</b>, plus two extras. The <i>ovector</i> argument is used in a
1716    different way, and this is described below. The other common arguments are used
1717    in the same way as for <b>pcre_exec()</b>, so their description is not repeated
1718    here.
1719    </P>
1720    <P>
1721    The two additional arguments provide workspace for the function. The workspace
1722    vector should contain at least 20 elements. It is used for keeping track of
1723    multiple paths through the pattern tree. More workspace will be needed for
1724    patterns and subjects where there are a lot of potential matches.
1725    </P>
1726    <P>
1727    Here is an example of a simple call to <b>pcre_dfa_exec()</b>:
1728    <pre>
1729      int rc;
1730      int ovector[10];
1731      int wspace[20];
1732      rc = pcre_dfa_exec(
1733        re,             /* result of pcre_compile() */
1734        NULL,           /* we didn't study the pattern */
1735        "some string",  /* the subject string */
1736        11,             /* the length of the subject string */
1737        0,              /* start at offset 0 in the subject */
1738        0,              /* default options */
1739        ovector,        /* vector of integers for substring information */
1740        10,             /* number of elements (NOT size in bytes) */
1741        wspace,         /* working space vector */
1742        20);            /* number of elements (NOT size in bytes) */
1743    </PRE>
1744    </P>
1745    <br><b>
1746    Option bits for <b>pcre_dfa_exec()</b>
1747    </b><br>
1748    <P>
1749    The unused bits of the <i>options</i> argument for <b>pcre_dfa_exec()</b> must be
1750    zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_<i>xxx</i>,
1751    PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NO_UTF8_CHECK, PCRE_PARTIAL,
1752    PCRE_DFA_SHORTEST, and PCRE_DFA_RESTART. All but the last three of these are
1753    the same as for <b>pcre_exec()</b>, so their description is not repeated here.
1754    <pre>
1755      PCRE_PARTIAL
1756    </pre>
1757    This has the same general effect as it does for <b>pcre_exec()</b>, but the
1758    details are slightly different. When PCRE_PARTIAL is set for
1759    <b>pcre_dfa_exec()</b>, the return code PCRE_ERROR_NOMATCH is converted into
1760    PCRE_ERROR_PARTIAL if the end of the subject is reached, there have been no
1761    complete matches, but there is still at least one matching possibility. The
1762    portion of the string that provided the partial match is set as the first
1763    matching string.
1764    <pre>
1765      PCRE_DFA_SHORTEST
1766    </pre>
1767    Setting the PCRE_DFA_SHORTEST option causes the matching algorithm to stop as
1768    soon as it has found one match. Because of the way the alternative algorithm
1769    works, this is necessarily the shortest possible match at the first possible
1770    matching point in the subject string.
1771    <pre>
1772      PCRE_DFA_RESTART
1773    </pre>
1774    When <b>pcre_dfa_exec()</b> is called with the PCRE_PARTIAL option, and returns
1775    a partial match, it is possible to call it again, with additional subject
1776    characters, and have it continue with the same match. The PCRE_DFA_RESTART
1777    option requests this action; when it is set, the <i>workspace</i> and
1778    <i>wscount</i> options must reference the same vector as before because data
1779    about the match so far is left in them after a partial match. There is more
1780    discussion of this facility in the
1781    <a href="pcrepartial.html"><b>pcrepartial</b></a>
1782    documentation.
1783    </P>
1784    <br><b>
1785    Successful returns from <b>pcre_dfa_exec()</b>
1786    </b><br>
1787    <P>
1788    When <b>pcre_dfa_exec()</b> succeeds, it may have matched more than one
1789    substring in the subject. Note, however, that all the matches from one run of
1790    the function start at the same point in the subject. The shorter matches are
1791    all initial substrings of the longer matches. For example, if the pattern
1792    <pre>
1793      &#60;.*&#62;
1794    </pre>
1795    is matched against the string
1796    <pre>
1797      This is &#60;something&#62; &#60;something else&#62; &#60;something further&#62; no more
1798    </pre>
1799    the three matched strings are
1800    <pre>
1801      &#60;something&#62;
1802      &#60;something&#62; &#60;something else&#62;
1803      &#60;something&#62; &#60;something else&#62; &#60;something further&#62;
1804    </pre>
1805    On success, the yield of the function is a number greater than zero, which is
1806    the number of matched substrings. The substrings themselves are returned in
1807    <i>ovector</i>. Each string uses two elements; the first is the offset to the
1808    start, and the second is the offset to the end. In fact, all the strings have
1809    the same start offset. (Space could have been saved by giving this only once,
1810    but it was decided to retain some compatibility with the way <b>pcre_exec()</b>
1811    returns data, even though the meaning of the strings is different.)
1812    </P>
1813    <P>
1814    The strings are returned in reverse order of length; that is, the longest
1815    matching string is given first. If there were too many matches to fit into
1816    <i>ovector</i>, the yield of the function is zero, and the vector is filled with
1817    the longest matches.
1818    </P>
1819    <br><b>
1820    Error returns from <b>pcre_dfa_exec()</b>
1821    </b><br>
1822    <P>
1823    The <b>pcre_dfa_exec()</b> function returns a negative number when it fails.
1824    Many of the errors are the same as for <b>pcre_exec()</b>, and these are
1825    described
1826    <a href="#errorlist">above.</a>
1827    There are in addition the following errors that are specific to
1828    <b>pcre_dfa_exec()</b>:
1829    <pre>
1830      PCRE_ERROR_DFA_UITEM      (-16)
1831    </pre>
1832    This return is given if <b>pcre_dfa_exec()</b> encounters an item in the pattern
1833    that it does not support, for instance, the use of \C or a back reference.
1834    <pre>
1835      PCRE_ERROR_DFA_UCOND      (-17)
1836    </pre>
1837    This return is given if <b>pcre_dfa_exec()</b> encounters a condition item that
1838    uses a back reference for the condition, or a test for recursion in a specific
1839    group. These are not supported.
1840    <pre>
1841      PCRE_ERROR_DFA_UMLIMIT    (-18)
1842    </pre>
1843    This return is given if <b>pcre_dfa_exec()</b> is called with an <i>extra</i>
1844    block that contains a setting of the <i>match_limit</i> field. This is not
1845    supported (it is meaningless).
1846    <pre>
1847      PCRE_ERROR_DFA_WSSIZE     (-19)
1848    </pre>
1849    This return is given if <b>pcre_dfa_exec()</b> runs out of space in the
1850    <i>workspace</i> vector.
1851    <pre>
1852      PCRE_ERROR_DFA_RECURSE    (-20)
1853    </pre>
1854    When a recursive subpattern is processed, the matching function calls itself
1855    recursively, using private vectors for <i>ovector</i> and <i>workspace</i>. This
1856    error is given if the output vector is not large enough. This should be
1857    extremely rare, as a vector of size 1000 is used.
1858    </P>
1859    <br><a name="SEC20" href="#TOC1">SEE ALSO</a><br>
1860    <P>
1861    <b>pcrebuild</b>(3), <b>pcrecallout</b>(3), <b>pcrecpp(3)</b>(3),
1862    <b>pcrematching</b>(3), <b>pcrepartial</b>(3), <b>pcreposix</b>(3),
1863    <b>pcreprecompile</b>(3), <b>pcresample</b>(3), <b>pcrestack</b>(3).
1864  </P>  </P>
1865    <br><a name="SEC21" href="#TOC1">AUTHOR</a><br>
1866  <P>  <P>
1867  Last updated: 03 February 2003  Philip Hazel
1868    <br>
1869    University Computing Service
1870    <br>
1871    Cambridge CB2 3QH, England.
1872    <br>
1873    </P>
1874    <br><a name="SEC22" href="#TOC1">REVISION</a><br>
1875    <P>
1876    Last updated: 04 June 2007
1877    <br>
1878    Copyright &copy; 1997-2007 University of Cambridge.
1879  <br>  <br>
1880  Copyright &copy; 1997-2003 University of Cambridge.  <p>
1881    Return to the <a href="index.html">PCRE index page</a>.
1882    </p>

Legend:
Removed from v.65  
changed lines
  Added in v.172

  ViewVC Help
Powered by ViewVC 1.1.5