/[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 142 by ph10, Fri Mar 30 15:55:18 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 four different conventions for indicating line breaks in
250    strings: a single CR (carriage return) character, a single LF (linefeed)
251    character, the two-character sequence CRLF, or any Unicode newline sequence.
252    The Unicode newline sequences are the three just mentioned, plus the single
253    characters VT (vertical tab, U+000B), FF (formfeed, U+000C), NEL (next line,
254    U+0085), LS (line separator, U+2028), and PS (paragraph separator, U+2029).
255    </P>
256    <P>
257    Each of the first three conventions is used by at least one operating system as
258    its standard newline sequence. When PCRE is built, a default can be specified.
259    The default default is LF, which is the Unix standard. When PCRE is run, the
260    default can be overridden, either when a pattern is compiled, or when it is
261    matched.
262    </P>
263    <P>
264    In the PCRE documentation the word "newline" is used to mean "the character or
265    pair of characters that indicate a line break". The choice of newline
266    convention affects the handling of the dot, circumflex, and dollar
267    metacharacters, the handling of #-comments in /x mode, and, when CRLF is a
268    recognized line ending sequence, the match position advancement for a
269    non-anchored pattern. The choice of newline convention does not affect the
270    interpretation of the \n or \r escape sequences.
271    </P>
272    <br><a name="SEC4" href="#TOC1">MULTITHREADING</a><br>
273  <P>  <P>
274  The PCRE functions can be used in multi-threading applications, with the  The PCRE functions can be used in multi-threading applications, with the
275  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>,
276  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
277  are shared by all threads.  callout function pointed to by <b>pcre_callout</b>, are shared by all threads.
278  </P>  </P>
279  <P>  <P>
280  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
281  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.
282  </P>  </P>
283  <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>
284    <P>
285    The compiled form of a regular expression can be saved and re-used at a later
286    time, possibly by a different program, and even on a host other than the one on
287    which it was compiled. Details are given in the
288    <a href="pcreprecompile.html"><b>pcreprecompile</b></a>
289    documentation.
290    </P>
291    <br><a name="SEC6" href="#TOC1">CHECKING BUILD-TIME OPTIONS</a><br>
292  <P>  <P>
293  <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>
294  </P>  </P>
# Line 186  documentation has more details about the Line 302  documentation has more details about the
302  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
303  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
304  which the information is placed. The following information is available:  which the information is placed. The following information is available:
 </P>  
 <P>  
305  <pre>  <pre>
306    PCRE_CONFIG_UTF8    PCRE_CONFIG_UTF8
307  </PRE>  </pre>
 </P>  
 <P>  
308  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;
309  otherwise it is set to zero.  otherwise it is set to zero.
310  </P>  <pre>
311  <P>    PCRE_CONFIG_UNICODE_PROPERTIES
312    </pre>
313    The output is an integer that is set to one if support for Unicode character
314    properties is available; otherwise it is set to zero.
315  <pre>  <pre>
316    PCRE_CONFIG_NEWLINE    PCRE_CONFIG_NEWLINE
317  </PRE>  </pre>
318  </P>  The output is an integer whose value specifies the default character sequence
319  <P>  that is recognized as meaning "newline". The four values that are supported
320  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, and -1 for ANY. The default should
321  the newline character. It is either linefeed (10) or carriage return (13), and  normally be the standard sequence for your operating system.
 should normally be the standard character for your operating system.  
 </P>  
 <P>  
322  <pre>  <pre>
323    PCRE_CONFIG_LINK_SIZE    PCRE_CONFIG_LINK_SIZE
324  </PRE>  </pre>
 </P>  
 <P>  
325  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
326  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
327  allow larger regular expressions to be compiled, at the expense of slower  allow larger regular expressions to be compiled, at the expense of slower
328  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
329  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>  
330  <pre>  <pre>
331    PCRE_CONFIG_POSIX_MALLOC_THRESHOLD    PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
332  </PRE>  </pre>
 </P>  
 <P>  
333  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
334  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
335  the <b>pcreposix</b> documentation.  the
336  </P>  <a href="pcreposix.html"><b>pcreposix</b></a>
337  <P>  documentation.
338  <pre>  <pre>
339    PCRE_CONFIG_MATCH_LIMIT    PCRE_CONFIG_MATCH_LIMIT
340  </PRE>  </pre>
 </P>  
 <P>  
341  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
342  internal matching function calls in a <b>pcre_exec()</b> execution. Further  internal matching function calls in a <b>pcre_exec()</b> execution. Further
343  details are given with <b>pcre_exec()</b> below.  details are given with <b>pcre_exec()</b> below.
344    <pre>
345      PCRE_CONFIG_MATCH_LIMIT_RECURSION
346    </pre>
347    The output is an integer that gives the default limit for the depth of
348    recursion when calling the internal matching function in a <b>pcre_exec()</b>
349    execution. Further details are given with <b>pcre_exec()</b> below.
350    <pre>
351      PCRE_CONFIG_STACKRECURSE
352    </pre>
353    The output is an integer that is set to one if internal recursion when running
354    <b>pcre_exec()</b> is implemented by recursive function calls that use the stack
355    to remember their state. This is the usual way that PCRE is compiled. The
356    output is zero if PCRE was compiled to use blocks of data on the heap instead
357    of recursive function calls. In this case, <b>pcre_stack_malloc</b> and
358    <b>pcre_stack_free</b> are called to manage memory blocks on the heap, thus
359    avoiding the use of the stack.
360  </P>  </P>
361  <br><a name="SEC5" href="#TOC1">COMPILING A PATTERN</a><br>  <br><a name="SEC7" href="#TOC1">COMPILING A PATTERN</a><br>
362  <P>  <P>
363  <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>
364  <b>const char **<i>errptr</i>, int *<i>erroffset</i>,</b>  <b>const char **<i>errptr</i>, int *<i>erroffset</i>,</b>
365  <b>const unsigned char *<i>tableptr</i>);</b>  <b>const unsigned char *<i>tableptr</i>);</b>
366    <b>pcre *pcre_compile2(const char *<i>pattern</i>, int <i>options</i>,</b>
367    <b>int *<i>errorcodeptr</i>,</b>
368    <b>const char **<i>errptr</i>, int *<i>erroffset</i>,</b>
369    <b>const unsigned char *<i>tableptr</i>);</b>
370  </P>  </P>
371  <P>  <P>
372  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
373  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
374  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,
375  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.
376  code and related data. The <b>pcre</b> type is defined for the returned block;  </P>
377  this is a typedef for a structure whose contents are not externally defined. It  <P>
378  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
379    <i>pattern</i> argument. A pointer to a single block of memory that is obtained
380    via <b>pcre_malloc</b> is returned. This contains the compiled code and related
381    data. The <b>pcre</b> type is defined for the returned block; this is a typedef
382    for a structure whose contents are not externally defined. It is up to the
383    caller to free the memory (via <b>pcre_free</b>) when it is no longer required.
384  </P>  </P>
385  <P>  <P>
386  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
387  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
388  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>
389  which is an address (see below).  argument, which is an address (see below).
390  </P>  </P>
391  <P>  <P>
392  The <i>options</i> argument contains independent bits that affect the  The <i>options</i> argument contains various bit settings that affect the
393  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
394  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
395  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
396  in the <b>pcrepattern</b> documentation). For these options, the contents of the  the detailed description in the
397  <i>options</i> argument specifies their initial settings at the start of  <a href="pcrepattern.html"><b>pcrepattern</b></a>
398  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
399    specifies their initial settings at the start of compilation and execution. The
400    PCRE_ANCHORED and PCRE_NEWLINE_<i>xxx</i> options can be set at the time of
401  matching as well as at compile time.  matching as well as at compile time.
402  </P>  </P>
403  <P>  <P>
404  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.
405  Otherwise, if compilation of a pattern fails, <b>pcre_compile()</b> returns  Otherwise, if compilation of a pattern fails, <b>pcre_compile()</b> returns
406  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
407  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
408  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
409    where the error was discovered is placed in the variable pointed to by
410  <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.
411  </P>  </P>
412  <P>  <P>
413  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
414  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
415  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
416  <b>pcre_maketables()</b>. See the section on locale support below.  textual error message. Error codes and messages are listed below.
417  </P>  </P>
418  <P>  <P>
419  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
420    character tables that are built when PCRE is compiled, using the default C
421    locale. Otherwise, <i>tableptr</i> must be an address that is the result of a
422    call to <b>pcre_maketables()</b>. This value is stored with the compiled
423    pattern, and used again by <b>pcre_exec()</b>, unless another table pointer is
424    passed to it. For more discussion, see the section on locale support below.
425  </P>  </P>
426  <P>  <P>
427    This code fragment shows a typical straightforward call to <b>pcre_compile()</b>:
428  <pre>  <pre>
429    pcre *re;    pcre *re;
430    const char *error;    const char *error;
# Line 297  This code fragment shows a typical strai Line 435  This code fragment shows a typical strai
435      &error,           /* for error message */      &error,           /* for error message */
436      &erroffset,       /* for error offset */      &erroffset,       /* for error offset */
437      NULL);            /* use default character tables */      NULL);            /* use default character tables */
438  </PRE>  </pre>
439  </P>  The following names for option bits are defined in the <b>pcre.h</b> header
440  <P>  file:
 The following option bits are defined:  
 </P>  
 <P>  
441  <pre>  <pre>
442    PCRE_ANCHORED    PCRE_ANCHORED
443  </PRE>  </pre>
 </P>  
 <P>  
444  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
445  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
446  being searched (the "subject string"). This effect can also be achieved by  being searched (the "subject string"). This effect can also be achieved by
447  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
448  Perl.  Perl.
449  </P>  <pre>
450  <P>    PCRE_AUTO_CALLOUT
451    </pre>
452    If this bit is set, <b>pcre_compile()</b> automatically inserts callout items,
453    all with number 255, before each pattern item. For discussion of the callout
454    facility, see the
455    <a href="pcrecallout.html"><b>pcrecallout</b></a>
456    documentation.
457  <pre>  <pre>
458    PCRE_CASELESS    PCRE_CASELESS
459  </PRE>  </pre>
 </P>  
 <P>  
460  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
461  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
462  pattern by a (?i) option setting.  pattern by a (?i) option setting. In UTF-8 mode, PCRE always understands the
463  </P>  concept of case for characters whose values are less than 128, so caseless
464  <P>  matching is always possible. For characters with higher values, the concept of
465    case is supported if PCRE is compiled with Unicode property support, but not
466    otherwise. If you want to use caseless matching for characters 128 and above,
467    you must ensure that PCRE is compiled with Unicode property support as well as
468    with UTF-8 support.
469  <pre>  <pre>
470    PCRE_DOLLAR_ENDONLY    PCRE_DOLLAR_ENDONLY
471  </PRE>  </pre>
 </P>  
 <P>  
472  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
473  end of the subject string. Without this option, a dollar also matches  end of the subject string. Without this option, a dollar also matches
474  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
475  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.
476  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
477  a pattern.  pattern.
 </P>  
 <P>  
478  <pre>  <pre>
479    PCRE_DOTALL    PCRE_DOTALL
480  </PRE>  </pre>
 </P>  
 <P>  
481  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,
482  including newlines. Without it, newlines are excluded. This option is  including those that indicate newline. Without it, a dot does not match when
483  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
484  (?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
485  character, independent of the setting of this option.  negative class such as [^a] always matches newline characters, independent of
486  </P>  the setting of this option.
487  <P>  <pre>
488      PCRE_DUPNAMES
489    </pre>
490    If this bit is set, names used to identify capturing subpatterns need not be
491    unique. This can be helpful for certain types of pattern when it is known that
492    only one instance of the named subpattern can ever be matched. There are more
493    details of named subpatterns below; see also the
494    <a href="pcrepattern.html"><b>pcrepattern</b></a>
495    documentation.
496  <pre>  <pre>
497    PCRE_EXTENDED    PCRE_EXTENDED
498  </PRE>  </pre>
 </P>  
 <P>  
499  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
500  ignored except when escaped or inside a character class. Whitespace does not  ignored except when escaped or inside a character class. Whitespace does not
501  include the VT character (code 11). In addition, characters between an  include the VT character (code 11). In addition, characters between an
502  unescaped # outside a character class and the next newline character,  unescaped # outside a character class and the next newline, inclusive, are also
503  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
504  be changed within a pattern by a (?x) option setting.  pattern by a (?x) option setting.
505  </P>  </P>
506  <P>  <P>
507  This option makes it possible to include comments inside complicated patterns.  This option makes it possible to include comments inside complicated patterns.
508  Note, however, that this applies only to data characters. Whitespace characters  Note, however, that this applies only to data characters. Whitespace characters
509  may never appear within special character sequences in a pattern, for example  may never appear within special character sequences in a pattern, for example
510  within the sequence (?( which introduces a conditional subpattern.  within the sequence (?( which introduces a conditional subpattern.
 </P>  
 <P>  
511  <pre>  <pre>
512    PCRE_EXTRA    PCRE_EXTRA
513  </PRE>  </pre>
 </P>  
 <P>  
514  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
515  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
516  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
517  special meaning causes an error, thus reserving these combinations for future  special meaning causes an error, thus reserving these combinations for future
518  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
519  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
520  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
521  pattern.  this option. It can also be set by a (?X) option setting within a pattern.
522  </P>  <pre>
523  <P>    PCRE_FIRSTLINE
524    </pre>
525    If this option is set, an unanchored pattern is required to match before or at
526    the first newline in the subject string, though the matched text may continue
527    over the newline.
528  <pre>  <pre>
529    PCRE_MULTILINE    PCRE_MULTILINE
530  </PRE>  </pre>
531  </P>  By default, PCRE treats the subject string as consisting of a single line of
532  <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"  
533  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
534  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
535  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 537  Perl.
537  </P>  </P>
538  <P>  <P>
539  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
540  match immediately following or immediately before any newline in the subject  match immediately following or immediately before internal newlines in the
541  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
542  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
543  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
544  occurrences of ^ or $ in a pattern, setting PCRE_MULTILINE has no effect.  occurrences of ^ or $ in a pattern, setting PCRE_MULTILINE has no effect.
545    <pre>
546      PCRE_NEWLINE_CR
547      PCRE_NEWLINE_LF
548      PCRE_NEWLINE_CRLF
549      PCRE_NEWLINE_ANY
550    </pre>
551    These options override the default newline definition that was chosen when PCRE
552    was built. Setting the first or the second specifies that a newline is
553    indicated by a single character (CR or LF, respectively). Setting
554    PCRE_NEWLINE_CRLF specifies that a newline is indicated by the two-character
555    CRLF sequence. Setting PCRE_NEWLINE_ANY specifies that any Unicode newline
556    sequence should be recognized. The Unicode newline sequences are the three just
557    mentioned, plus the single characters VT (vertical tab, U+000B), FF (formfeed,
558    U+000C), NEL (next line, U+0085), LS (line separator, U+2028), and PS
559    (paragraph separator, U+2029). The last two are recognized only in UTF-8 mode.
560    </P>
561    <P>
562    The newline setting in the options word uses three bits that are treated
563    as a number, giving eight possibilities. Currently only five are used (default
564    plus the four values above). This means that if you set more than one newline
565    option, the combination may or may not be sensible. For example,
566    PCRE_NEWLINE_CR with PCRE_NEWLINE_LF is equivalent to PCRE_NEWLINE_CRLF, but
567    other combinations yield unused numbers and cause an error.
568    </P>
569    <P>
570    The only time that a line break is specially recognized when compiling a
571    pattern is if PCRE_EXTENDED is set, and an unescaped # outside a character
572    class is encountered. This indicates a comment that lasts until after the next
573    line break sequence. In other circumstances, line break sequences are treated
574    as literal data, except that in PCRE_EXTENDED mode, both CR and LF are treated
575    as whitespace characters and are therefore ignored.
576  </P>  </P>
577  <P>  <P>
578    The newline option that is set at compile time becomes the default that is used
579    for <b>pcre_exec()</b> and <b>pcre_dfa_exec()</b>, but it can be overridden.
580  <pre>  <pre>
581    PCRE_NO_AUTO_CAPTURE    PCRE_NO_AUTO_CAPTURE
582  </PRE>  </pre>
 </P>  
 <P>  
583  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
584  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
585  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
586  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
587  in Perl.  in Perl.
 </P>  
 <P>  
588  <pre>  <pre>
589    PCRE_UNGREEDY    PCRE_UNGREEDY
590  </PRE>  </pre>
 </P>  
 <P>  
591  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
592  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
593  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>  
594  <pre>  <pre>
595    PCRE_UTF8    PCRE_UTF8
596  </PRE>  </pre>
 </P>  
 <P>  
597  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
598  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
599  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
600  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
601  behaviour of PCRE are given in the  behaviour of PCRE are given in the
602  <a href="pcre.html#utf8support">section on UTF-8 support</a>  <a href="pcre.html#utf8support">section on UTF-8 support</a>
603  in the main  in the main
604  <a href="pcre.html"><b>pcre</b></a>  <a href="pcre.html"><b>pcre</b></a>
605  page.  page.
606    <pre>
607      PCRE_NO_UTF8_CHECK
608    </pre>
609    When PCRE_UTF8 is set, the validity of the pattern as a UTF-8 string is
610    automatically checked. If an invalid UTF-8 sequence of bytes is found,
611    <b>pcre_compile()</b> returns an error. If you already know that your pattern is
612    valid, and you want to skip this check for performance reasons, you can set the
613    PCRE_NO_UTF8_CHECK option. When it is set, the effect of passing an invalid
614    UTF-8 string as a pattern is undefined. It may cause your program to crash.
615    Note that this option can also be passed to <b>pcre_exec()</b> and
616    <b>pcre_dfa_exec()</b>, to suppress the UTF-8 validity checking of subject
617    strings.
618    </P>
619    <br><a name="SEC8" href="#TOC1">COMPILATION ERROR CODES</a><br>
620    <P>
621    The following table lists the error codes than may be returned by
622    <b>pcre_compile2()</b>, along with the error messages that may be returned by
623    both compiling functions. As PCRE has developed, some error codes have fallen
624    out of use. To avoid confusion, they have not been re-used.
625    <pre>
626       0  no error
627       1  \ at end of pattern
628       2  \c at end of pattern
629       3  unrecognized character follows \
630       4  numbers out of order in {} quantifier
631       5  number too big in {} quantifier
632       6  missing terminating ] for character class
633       7  invalid escape sequence in character class
634       8  range out of order in character class
635       9  nothing to repeat
636      10  [this code is not in use]
637      11  internal error: unexpected repeat
638      12  unrecognized character after (?
639      13  POSIX named classes are supported only within a class
640      14  missing )
641      15  reference to non-existent subpattern
642      16  erroffset passed as NULL
643      17  unknown option bit(s) set
644      18  missing ) after comment
645      19  [this code is not in use]
646      20  regular expression too large
647      21  failed to get memory
648      22  unmatched parentheses
649      23  internal error: code overflow
650      24  unrecognized character after (?&#60;
651      25  lookbehind assertion is not fixed length
652      26  malformed number or name after (?(
653      27  conditional group contains more than two branches
654      28  assertion expected after (?(
655      29  (?R or (?digits must be followed by )
656      30  unknown POSIX class name
657      31  POSIX collating elements are not supported
658      32  this version of PCRE is not compiled with PCRE_UTF8 support
659      33  [this code is not in use]
660      34  character value in \x{...} sequence is too large
661      35  invalid condition (?(0)
662      36  \C not allowed in lookbehind assertion
663      37  PCRE does not support \L, \l, \N, \U, or \u
664      38  number after (?C is &#62; 255
665      39  closing ) for (?C expected
666      40  recursive call could loop indefinitely
667      41  unrecognized character after (?P
668      42  syntax error in subpattern name (missing terminator)
669      43  two named subpatterns have the same name
670      44  invalid UTF-8 string
671      45  support for \P, \p, and \X has not been compiled
672      46  malformed \P or \p sequence
673      47  unknown property name after \P or \p
674      48  subpattern name is too long (maximum 32 characters)
675      49  too many named subpatterns (maximum 10,000)
676      50  repeated subpattern is too long
677      51  octal value is greater than \377 (not in UTF-8 mode)
678      52  internal error: overran compiling workspace
679      53  internal error: previously-checked referenced subpattern not found
680      54  DEFINE group contains more than one branch
681      55  repeating a DEFINE group is not allowed
682      56  inconsistent NEWLINE options"
683    </PRE>
684  </P>  </P>
685  <br><a name="SEC6" href="#TOC1">STUDYING A PATTERN</a><br>  <br><a name="SEC9" href="#TOC1">STUDYING A PATTERN</a><br>
686  <P>  <P>
687  <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>
688  <b>const char **<i>errptr</i>);</b>  <b>const char **<i>errptr</i>);</b>
689  </P>  </P>
690  <P>  <P>
691  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
692  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
693  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
694  argument. If studing the pattern produces additional information that will help  argument. If studying the pattern produces additional information that will
695  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
696  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
697    results of the study.
698  </P>  </P>
699  <P>  <P>
700  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
701  <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
702  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
703  described below. If studying the pattern does not produce any additional  described
704  information, <b>pcre_study()</b> returns NULL. In that circumstance, if the  <a href="#extradata">below</a>
705  calling program wants to pass some of the other fields to <b>pcre_exec()</b>, it  in the section on matching a pattern.
706  must set up its own <b>pcre_extra</b> block.  </P>
707    <P>
708    If studying the pattern does not produce any additional information
709    <b>pcre_study()</b> returns NULL. In that circumstance, if the calling program
710    wants to pass any of the other fields to <b>pcre_exec()</b>, it must set up its
711    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  The second argument of <b>pcre_study()</b> contains option bits. At present, no
715  for <b>pcre_study()</b>, and this argument should always be zero.  options are defined, and this argument should always be zero.
716  </P>  </P>
717  <P>  <P>
718  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
719  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
720  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
721  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
722  be sure that it has run successfully.  should test the error pointer for NULL after calling <b>pcre_study()</b>, to be
723    sure that it has run successfully.
724  </P>  </P>
725  <P>  <P>
726  This is a typical call to <b>pcre_study</b>():  This is a typical call to <b>pcre_study</b>():
 </P>  
 <P>  
727  <pre>  <pre>
728    pcre_extra *pe;    pcre_extra *pe;
729    pe = pcre_study(    pe = pcre_study(
730      re,             /* result of pcre_compile() */      re,             /* result of pcre_compile() */
731      0,              /* no options exist */      0,              /* no options exist */
732      &error);        /* set to NULL or points to a message */      &error);        /* set to NULL or points to a message */
733  </PRE>  </pre>
 </P>  
 <P>  
734  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
735  not have a single fixed starting character. A bitmap of possible starting  not have a single fixed starting character. A bitmap of possible starting
736  characters is created.  bytes is created.
737  </P>  <a name="localesupport"></a></P>
738  <a name="localesupport"></a><br><a name="SEC7" href="#TOC1">LOCALE SUPPORT</a><br>  <br><a name="SEC10" href="#TOC1">LOCALE SUPPORT</a><br>
739  <P>  <P>
740  PCRE handles caseless matching, and determines whether characters are letters,  PCRE handles caseless matching, and determines whether characters are letters,
741  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
742  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
743  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
744  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
745  is NULL, and is sufficient for many applications.  support. The use of locales with Unicode is discouraged. If you are handling
746  </P>  characters with codes greater than 128, you should either use UTF-8 and
747  <P>  Unicode, or use locales, but not try to mix the two.
748  An alternative set of tables can, however, be supplied. Such tables are built  </P>
749  by calling the <b>pcre_maketables()</b> function, which has no arguments, in the  <P>
750  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
751  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.
752  French locale (where accented characters with codes greater than 128 are  Normally, the internal tables recognize only ASCII characters. However, when
753  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
754  </P>  default "C" locale of the local system, which may cause them to be different.
755  <P>  </P>
756    <P>
757    The internal tables can always be overridden by tables supplied by the
758    application that calls PCRE. These may be created in a different locale from
759    the default. As more and more applications change to using Unicode, the need
760    for this locale support is expected to die away.
761    </P>
762    <P>
763    External tables are built by calling the <b>pcre_maketables()</b> function,
764    which has no arguments, in the relevant locale. The result can then be passed
765    to <b>pcre_compile()</b> or <b>pcre_exec()</b> as often as necessary. For
766    example, to build and use tables that are appropriate for the French locale
767    (where accented characters with values greater than 128 are treated as letters),
768    the following code could be used:
769  <pre>  <pre>
770    setlocale(LC_CTYPE, "fr");    setlocale(LC_CTYPE, "fr_FR");
771    tables = pcre_maketables();    tables = pcre_maketables();
772    re = pcre_compile(..., tables);    re = pcre_compile(..., tables);
773  </PRE>  </pre>
774    The locale name "fr_FR" is used on Linux and other Unix-like systems; if you
775    are using Windows, the name for the French locale is "french".
776    </P>
777    <P>
778    When <b>pcre_maketables()</b> runs, the tables are built in memory that is
779    obtained via <b>pcre_malloc</b>. It is the caller's responsibility to ensure
780    that the memory containing the tables remains available for as long as it is
781    needed.
782  </P>  </P>
783  <P>  <P>
784  The tables are built in memory that is obtained via <b>pcre_malloc</b>. The  The pointer that is passed to <b>pcre_compile()</b> is saved with the compiled
 pointer that is passed to <b>pcre_compile</b> is saved with the compiled  
785  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>
786  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
787  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
788  in different locales. It is the caller's responsibility to ensure that the  different patterns can be compiled in different locales.
 memory containing the tables remains available for as long as it is needed.  
789  </P>  </P>
790  <br><a name="SEC8" href="#TOC1">INFORMATION ABOUT A PATTERN</a><br>  <P>
791    It is possible to pass a table pointer or NULL (indicating the use of the
792    internal tables) to <b>pcre_exec()</b>. Although not intended for this purpose,
793    this facility could be used to match a pattern in a different locale from the
794    one in which it was compiled. Passing table pointers at run time is discussed
795    below in the section on matching a pattern.
796    </P>
797    <br><a name="SEC11" href="#TOC1">INFORMATION ABOUT A PATTERN</a><br>
798  <P>  <P>
799  <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>
800  <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 811  the pattern was not studied. The third a
811  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
812  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
813  the following negative numbers:  the following negative numbers:
 </P>  
 <P>  
814  <pre>  <pre>
815    PCRE_ERROR_NULL       the argument <i>code</i> was NULL    PCRE_ERROR_NULL       the argument <i>code</i> was NULL
816                          the argument <i>where</i> was NULL                          the argument <i>where</i> was NULL
817    PCRE_ERROR_BADMAGIC   the "magic number" was not found    PCRE_ERROR_BADMAGIC   the "magic number" was not found
818    PCRE_ERROR_BADOPTION  the value of <i>what</i> was invalid    PCRE_ERROR_BADOPTION  the value of <i>what</i> was invalid
819  </PRE>  </pre>
820  </P>  The "magic number" is placed at the start of each compiled pattern as an simple
821  <P>  check against passing an arbitrary memory pointer. Here is a typical call of
822  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>  
823  <pre>  <pre>
824    int rc;    int rc;
825    unsigned long int length;    size_t length;
826    rc = pcre_fullinfo(    rc = pcre_fullinfo(
827      re,               /* result of pcre_compile() */      re,               /* result of pcre_compile() */
828      pe,               /* result of pcre_study(), or NULL */      pe,               /* result of pcre_study(), or NULL */
829      PCRE_INFO_SIZE,   /* what is required */      PCRE_INFO_SIZE,   /* what is required */
830      &length);         /* where to put the data */      &length);         /* where to put the data */
831  </PRE>  </pre>
 </P>  
 <P>  
832  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
833  as follows:  as follows:
 </P>  
 <P>  
834  <pre>  <pre>
835    PCRE_INFO_BACKREFMAX    PCRE_INFO_BACKREFMAX
836  </PRE>  </pre>
 </P>  
 <P>  
837  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
838  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
839  no back references.  no back references.
 </P>  
 <P>  
840  <pre>  <pre>
841    PCRE_INFO_CAPTURECOUNT    PCRE_INFO_CAPTURECOUNT
842  </PRE>  </pre>
 </P>  
 <P>  
843  Return the number of capturing subpatterns in the pattern. The fourth argument  Return the number of capturing subpatterns in the pattern. The fourth argument
844  should point to an \fbint\fR variable.  should point to an <b>int</b> variable.
845  </P>  <pre>
846  <P>    PCRE_INFO_DEFAULT_TABLES
847    </pre>
848    Return a pointer to the internal default character tables within PCRE. The
849    fourth argument should point to an <b>unsigned char *</b> variable. This
850    information call is provided for internal use by the <b>pcre_study()</b>
851    function. External callers can cause PCRE to use its internal tables by passing
852    a NULL table pointer.
853  <pre>  <pre>
854    PCRE_INFO_FIRSTBYTE    PCRE_INFO_FIRSTBYTE
855  </PRE>  </pre>
 </P>  
 <P>  
856  Return information about the first byte of any matched string, for a  Return information about the first byte of any matched string, for a
857  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>
858  old name is still recognized for backwards compatibility.)  variable. (This option used to be called PCRE_INFO_FIRSTCHAR; the old name is
859  </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  
860  </P>  </P>
861  <P>  <P>
862    If there is a fixed first byte, for example, from a pattern such as
863    (cat|cow|coyote), its value is returned. Otherwise, if either
864    <br>
865    <br>
866  (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
867  starts with "^", or  starts with "^", or
868  </P>  <br>
869  <P>  <br>
870  (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
871  (if it were set, the pattern would be anchored),  (if it were set, the pattern would be anchored),
872  </P>  <br>
873  <P>  <br>
874  -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
875  subject string or after any newline within the string. Otherwise -2 is  subject string or after any newline within the string. Otherwise -2 is
876  returned. For anchored patterns, -2 is returned.  returned. For anchored patterns, -2 is returned.
 </P>  
 <P>  
877  <pre>  <pre>
878    PCRE_INFO_FIRSTTABLE    PCRE_INFO_FIRSTTABLE
879  </PRE>  </pre>
 </P>  
 <P>  
880  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
881  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
882  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
883  fourth argument should point to an <b>unsigned char *</b> variable.  fourth argument should point to an <b>unsigned char *</b> variable.
 </P>  
 <P>  
884  <pre>  <pre>
885    PCRE_INFO_LASTLITERAL    PCRE_INFO_LASTLITERAL
886  </PRE>  </pre>
 </P>  
 <P>  
887  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
888  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
889  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 891  returned. For anchored patterns, a last
891  follows something of variable length. For example, for the pattern  follows something of variable length. For example, for the pattern
892  /^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
893  is -1.  is -1.
 </P>  
 <P>  
894  <pre>  <pre>
895    PCRE_INFO_NAMECOUNT    PCRE_INFO_NAMECOUNT
896    PCRE_INFO_NAMEENTRYSIZE    PCRE_INFO_NAMEENTRYSIZE
897    PCRE_INFO_NAMETABLE    PCRE_INFO_NAMETABLE
898  </PRE>  </pre>
 </P>  
 <P>  
899  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
900  names are just an additional way of identifying the parentheses, which still  names are just an additional way of identifying the parentheses, which still
901  acquire a number. A caller that wants to extract data from a named subpattern  acquire numbers. Several convenience functions such as
902  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
903  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
904  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
905  table for the pattern.  output vector (described with <b>pcre_exec()</b> below). To do the conversion,
906    you need to use the name-to-number map, which is described by these three
907    values.
908  </P>  </P>
909  <P>  <P>
910  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 914  length of the longest name. PCRE_INFO_NA
914  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
915  are the number of the capturing parenthesis, most significant byte first. The  are the number of the capturing parenthesis, most significant byte first. The
916  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
917  alphabetical order. For example, consider the following pattern (assume  alphabetical order. When PCRE_DUPNAMES is set, duplicate names are in order of
918    their parentheses numbers. For example, consider the following pattern (assume
919  PCRE_EXTENDED is set, so white space - including newlines - is ignored):  PCRE_EXTENDED is set, so white space - including newlines - is ignored):
 </P>  
 <P>  
920  <pre>  <pre>
921    (?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) )
922    (?P&#60;month&#62;\d\d) - (?P&#60;day&#62;\d\d) )  </pre>
 </PRE>  
 </P>  
 <P>  
923  There are four named subpatterns, so the table has four entries, and each entry  There are four named subpatterns, so the table has four entries, and each entry
924  in the table is eight bytes long. The table is as follows, with non-printing  in the table is eight bytes long. The table is as follows, with non-printing
925  bytes shows in hex, and undefined bytes shown as ??:  bytes shows in hexadecimal, and undefined bytes shown as ??:
 </P>  
 <P>  
926  <pre>  <pre>
927    00 01 d  a  t  e  00 ??    00 01 d  a  t  e  00 ??
928    00 05 d  a  y  00 ?? ??    00 05 d  a  y  00 ?? ??
929    00 04 m  o  n  t  h  00    00 04 m  o  n  t  h  00
930    00 02 y  e  a  r  00 ??    00 02 y  e  a  r  00 ??
931  </PRE>  </pre>
932  </P>  When writing code to extract data from named subpatterns using the
933  <P>  name-to-number map, remember that the length of the entries is likely to be
934  When writing code to extract data from named subpatterns, remember that the  different for each compiled pattern.
 length of each entry may be different for each compiled pattern.  
 </P>  
 <P>  
935  <pre>  <pre>
936    PCRE_INFO_OPTIONS    PCRE_INFO_OPTIONS
937  </PRE>  </pre>
 </P>  
 <P>  
938  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
939  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
940  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 943  top-level option settings within the pat
943  <P>  <P>
944  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
945  alternatives begin with one of the following:  alternatives begin with one of the following:
 </P>  
 <P>  
946  <pre>  <pre>
947    ^     unless PCRE_MULTILINE is set    ^     unless PCRE_MULTILINE is set
948    \A    always    \A    always
949    \G    always    \G    always
950    .*    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
951            references to the subpattern in which .* appears  </pre>
 </PRE>  
 </P>  
 <P>  
952  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
953  <b>pcre_fullinfo()</b>.  <b>pcre_fullinfo()</b>.
 </P>  
 <P>  
954  <pre>  <pre>
955    PCRE_INFO_SIZE    PCRE_INFO_SIZE
956  </PRE>  </pre>
 </P>  
 <P>  
957  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
958  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
959  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>
960  variable.  variable.
 </P>  
 <P>  
961  <pre>  <pre>
962    PCRE_INFO_STUDYSIZE    PCRE_INFO_STUDYSIZE
963  </PRE>  </pre>
964  </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  
965  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
966  <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
967  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
968  <b>size_t</b> variable.  <b>size_t</b> variable.
969  </P>  </P>
970  <br><a name="SEC9" href="#TOC1">OBSOLETE INFO FUNCTION</a><br>  <br><a name="SEC12" href="#TOC1">OBSOLETE INFO FUNCTION</a><br>
971  <P>  <P>
972  <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>
973  <b>*<i>firstcharptr</i>);</b>  <b>*<i>firstcharptr</i>);</b>
# Line 753  restrictive to return all the available Line 978  restrictive to return all the available
978  programs should use <b>pcre_fullinfo()</b> instead. The yield of  programs should use <b>pcre_fullinfo()</b> instead. The yield of
979  <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
980  following negative numbers:  following negative numbers:
 </P>  
 <P>  
981  <pre>  <pre>
982    PCRE_ERROR_NULL       the argument <i>code</i> was NULL    PCRE_ERROR_NULL       the argument <i>code</i> was NULL
983    PCRE_ERROR_BADMAGIC   the "magic number" was not found    PCRE_ERROR_BADMAGIC   the "magic number" was not found
984  </PRE>  </pre>
 </P>  
 <P>  
985  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
986  pattern was compiled is placed in the integer it points to (see  pattern was compiled is placed in the integer it points to (see
987  PCRE_INFO_OPTIONS above).  PCRE_INFO_OPTIONS above).
# Line 770  If the pattern is not anchored and the < Line 991  If the pattern is not anchored and the <
991  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
992  string (see PCRE_INFO_FIRSTBYTE above).  string (see PCRE_INFO_FIRSTBYTE above).
993  </P>  </P>
994  <br><a name="SEC10" href="#TOC1">MATCHING A PATTERN</a><br>  <br><a name="SEC13" href="#TOC1">REFERENCE COUNTS</a><br>
995    <P>
996    <b>int pcre_refcount(pcre *<i>code</i>, int <i>adjust</i>);</b>
997    </P>
998    <P>
999    The <b>pcre_refcount()</b> function is used to maintain a reference count in the
1000    data block that contains a compiled pattern. It is provided for the benefit of
1001    applications that operate in an object-oriented manner, where different parts
1002    of the application may be using the same compiled pattern, but you want to free
1003    the block when they are all done.
1004    </P>
1005    <P>
1006    When a pattern is compiled, the reference count field is initialized to zero.
1007    It is changed only by calling this function, whose action is to add the
1008    <i>adjust</i> value (which may be positive or negative) to it. The yield of the
1009    function is the new value. However, the value of the count is constrained to
1010    lie between 0 and 65535, inclusive. If the new value is outside these limits,
1011    it is forced to the appropriate limit value.
1012    </P>
1013    <P>
1014    Except when it is zero, the reference count is not correctly preserved if a
1015    pattern is compiled on one host and then transferred to a host whose byte-order
1016    is different. (This seems a highly unlikely scenario.)
1017    </P>
1018    <br><a name="SEC14" href="#TOC1">MATCHING A PATTERN: THE TRADITIONAL FUNCTION</a><br>
1019  <P>  <P>
1020  <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>
1021  <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 1023  string (see PCRE_INFO_FIRSTBYTE above).
1023  </P>  </P>
1024  <P>  <P>
1025  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
1026  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
1027  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
1028  <i>extra</i> argument.  <i>extra</i> argument. This function is the main matching facility of the
1029    library, and it operates in a Perl-like manner. For specialist use there is
1030    also an alternative matching function, which is described
1031    <a href="#dfamatch">below</a>
1032    in the section about the <b>pcre_dfa_exec()</b> function.
1033  </P>  </P>
1034  <P>  <P>
1035  Here is an example of a simple call to <b>pcre_exec()</b>:  In most applications, the pattern will have been compiled (and optionally
1036    studied) in the same process that calls <b>pcre_exec()</b>. However, it is
1037    possible to save compiled patterns and study data, and then use them later
1038    in different processes, possibly even on different hosts. For a discussion
1039    about this, see the
1040    <a href="pcreprecompile.html"><b>pcreprecompile</b></a>
1041    documentation.
1042  </P>  </P>
1043  <P>  <P>
1044    Here is an example of a simple call to <b>pcre_exec()</b>:
1045  <pre>  <pre>
1046    int rc;    int rc;
1047    int ovector[30];    int ovector[30];
# Line 796  Here is an example of a simple call to < Line 1052  Here is an example of a simple call to <
1052      11,             /* the length of the subject string */      11,             /* the length of the subject string */
1053      0,              /* start at offset 0 in the subject */      0,              /* start at offset 0 in the subject */
1054      0,              /* default options */      0,              /* default options */
1055      ovector,        /* vector for substring information */      ovector,        /* vector of integers for substring information */
1056      30);            /* number of elements in the vector */      30);            /* number of elements (NOT size in bytes) */
1057  </PRE>  <a name="extradata"></a></PRE>
1058  </P>  </P>
1059    <br><b>
1060    Extra data for <b>pcre_exec()</b>
1061    </b><br>
1062  <P>  <P>
1063  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>
1064  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
1065  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
1066  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
1067  </P>  fields (not necessarily in this order):
 <P>  
1068  <pre>  <pre>
1069    unsigned long int <i>flags</i>;    unsigned long int <i>flags</i>;
1070    void *<i>study_data</i>;    void *<i>study_data</i>;
1071    unsigned long int <i>match_limit</i>;    unsigned long int <i>match_limit</i>;
1072      unsigned long int <i>match_limit_recursion</i>;
1073    void *<i>callout_data</i>;    void *<i>callout_data</i>;
1074  </PRE>    const unsigned char *<i>tables</i>;
1075  </P>  </pre>
 <P>  
1076  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
1077  are set. The flag bits are:  are set. The flag bits are:
 </P>  
 <P>  
1078  <pre>  <pre>
1079    PCRE_EXTRA_STUDY_DATA    PCRE_EXTRA_STUDY_DATA
1080    PCRE_EXTRA_MATCH_LIMIT    PCRE_EXTRA_MATCH_LIMIT
1081      PCRE_EXTRA_MATCH_LIMIT_RECURSION
1082    PCRE_EXTRA_CALLOUT_DATA    PCRE_EXTRA_CALLOUT_DATA
1083  </PRE>    PCRE_EXTRA_TABLES
1084  </P>  </pre>
 <P>  
1085  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
1086  <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
1087  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
1088  the block by setting the other fields.  the block by setting the other fields and their corresponding flag bits.
1089  </P>  </P>
1090  <P>  <P>
1091  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
1092  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,
1093  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
1094  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.  
1095  </P>  </P>
1096  <P>  <P>
1097  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
1098  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
1099  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
1100  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
1101  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
1102  exceeded, <b>pcre_exec()</b> returns PCRE_ERROR_MATCHLIMIT.  string.
1103  </P>  </P>
1104  <P>  <P>
1105  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
1106  which is described in the <b>pcrecallout</b> documentation.  default is 10 million, which handles all but the most extreme cases. You can
1107  </P>  override the default by suppling <b>pcre_exec()</b> with a <b>pcre_extra</b>
1108  <P>  block in which <i>match_limit</i> is set, and PCRE_EXTRA_MATCH_LIMIT is set in
1109  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
1110  unused bits must be zero. This limits <b>pcre_exec()</b> to matching at the  PCRE_ERROR_MATCHLIMIT.
1111  first matching position. However, if a pattern was compiled with PCRE_ANCHORED,  </P>
1112  or turned out to be anchored by virtue of its contents, it cannot be made  <P>
1113  unachored at matching time.  The <i>match_limit_recursion</i> field is similar to <i>match_limit</i>, but
1114    instead of limiting the total number of times that <b>match()</b> is called, it
1115    limits the depth of recursion. The recursion depth is a smaller number than the
1116    total number of calls, because not all calls to <b>match()</b> are recursive.
1117    This limit is of use only if it is set smaller than <i>match_limit</i>.
1118    </P>
1119    <P>
1120    Limiting the recursion depth limits the amount of stack that can be used, or,
1121    when PCRE has been compiled to use memory on the heap instead of the stack, the
1122    amount of heap memory that can be used.
1123    </P>
1124    <P>
1125    The default value for <i>match_limit_recursion</i> can be set when PCRE is
1126    built; the default default is the same value as the default for
1127    <i>match_limit</i>. You can override the default by suppling <b>pcre_exec()</b>
1128    with a <b>pcre_extra</b> block in which <i>match_limit_recursion</i> is set, and
1129    PCRE_EXTRA_MATCH_LIMIT_RECURSION is set in the <i>flags</i> field. If the limit
1130    is exceeded, <b>pcre_exec()</b> returns PCRE_ERROR_RECURSIONLIMIT.
1131  </P>  </P>
1132  <P>  <P>
1133  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,
1134    which is described in the
1135    <a href="pcrecallout.html"><b>pcrecallout</b></a>
1136    documentation.
1137  </P>  </P>
1138  <P>  <P>
1139    The <i>tables</i> field is used to pass a character tables pointer to
1140    <b>pcre_exec()</b>; this overrides the value that is stored with the compiled
1141    pattern. A non-NULL value is stored with the compiled pattern only if custom
1142    tables were supplied to <b>pcre_compile()</b> via its <i>tableptr</i> argument.
1143    If NULL is passed to <b>pcre_exec()</b> using this mechanism, it forces PCRE's
1144    internal tables to be used. This facility is helpful when re-using patterns
1145    that have been saved after compiling with an external set of tables, because
1146    the external tables might be at a different address when <b>pcre_exec()</b> is
1147    called. See the
1148    <a href="pcreprecompile.html"><b>pcreprecompile</b></a>
1149    documentation for a discussion of saving compiled patterns for later use.
1150    </P>
1151    <br><b>
1152    Option bits for <b>pcre_exec()</b>
1153    </b><br>
1154    <P>
1155    The unused bits of the <i>options</i> argument for <b>pcre_exec()</b> must be
1156    zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_<i>xxx</i>,
1157    PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NO_UTF8_CHECK and PCRE_PARTIAL.
1158    <pre>
1159      PCRE_ANCHORED
1160    </pre>
1161    The PCRE_ANCHORED option limits <b>pcre_exec()</b> to matching at the first
1162    matching position. If a pattern was compiled with PCRE_ANCHORED, or turned out
1163    to be anchored by virtue of its contents, it cannot be made unachored at
1164    matching time.
1165    <pre>
1166      PCRE_NEWLINE_CR
1167      PCRE_NEWLINE_LF
1168      PCRE_NEWLINE_CRLF
1169      PCRE_NEWLINE_ANY
1170    </pre>
1171    These options override the newline definition that was chosen or defaulted when
1172    the pattern was compiled. For details, see the description of
1173    <b>pcre_compile()</b> above. During matching, the newline choice affects the
1174    behaviour of the dot, circumflex, and dollar metacharacters. It may also alter
1175    the way the match position is advanced after a match failure for an unanchored
1176    pattern. When PCRE_NEWLINE_CRLF or PCRE_NEWLINE_ANY is set, and a match attempt
1177    fails when the current position is at a CRLF sequence, the match position is
1178    advanced by two characters instead of one, in other words, to after the CRLF.
1179  <pre>  <pre>
1180    PCRE_NOTBOL    PCRE_NOTBOL
1181  </PRE>  </pre>
1182  </P>  This option specifies that first character of the subject string is not the
1183  <P>  beginning of a line, so the circumflex metacharacter should not match before
1184  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
1185  circumflex metacharacter should not match before it. Setting this without  never to match. This option affects only the behaviour of the circumflex
1186  PCRE_MULTILINE (at compile time) causes circumflex never to match.  metacharacter. It does not affect \A.
 </P>  
 <P>  
1187  <pre>  <pre>
1188    PCRE_NOTEOL    PCRE_NOTEOL
1189  </PRE>  </pre>
1190  </P>  This option specifies that the end of the subject string is not the end of a
1191  <P>  line, so the dollar metacharacter should not match it nor (except in multiline
1192  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
1193  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
1194  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>  
1195  <pre>  <pre>
1196    PCRE_NOTEMPTY    PCRE_NOTEMPTY
1197  </PRE>  </pre>
 </P>  
 <P>  
1198  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
1199  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
1200  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>  
1201  <pre>  <pre>
1202    a?b?    a?b?
1203  </PRE>  </pre>
 </P>  
 <P>  
1204  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
1205  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
1206  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 1210  Perl has no direct equivalent of PCRE_NO
1210  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
1211  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
1212  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
1213  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
1214  below) and trying an ordinary match again.  starting offset (see below) and trying an ordinary match again. There is some
1215    code that demonstrates how to do this in the <i>pcredemo.c</i> sample program.
1216    <pre>
1217      PCRE_NO_UTF8_CHECK
1218    </pre>
1219    When PCRE_UTF8 is set at compile time, the validity of the subject as a UTF-8
1220    string is automatically checked when <b>pcre_exec()</b> is subsequently called.
1221    The value of <i>startoffset</i> is also checked to ensure that it points to the
1222    start of a UTF-8 character. If an invalid UTF-8 sequence of bytes is found,
1223    <b>pcre_exec()</b> returns the error PCRE_ERROR_BADUTF8. If <i>startoffset</i>
1224    contains an invalid value, PCRE_ERROR_BADUTF8_OFFSET is returned.
1225    </P>
1226    <P>
1227    If you already know that your subject is valid, and you want to skip these
1228    checks for performance reasons, you can set the PCRE_NO_UTF8_CHECK option when
1229    calling <b>pcre_exec()</b>. You might want to do this for the second and
1230    subsequent calls to <b>pcre_exec()</b> if you are making repeated calls to find
1231    all the matches in a single subject string. However, you should be sure that
1232    the value of <i>startoffset</i> points to the start of a UTF-8 character. When
1233    PCRE_NO_UTF8_CHECK is set, the effect of passing an invalid UTF-8 string as a
1234    subject, or a value of <i>startoffset</i> that does not point to the start of a
1235    UTF-8 character, is undefined. Your program may crash.
1236    <pre>
1237      PCRE_PARTIAL
1238    </pre>
1239    This option turns on the partial matching feature. If the subject string fails
1240    to match the pattern, but at some point during the matching process the end of
1241    the subject was reached (that is, the subject partially matches the pattern and
1242    the failure to match occurred only because there were not enough subject
1243    characters), <b>pcre_exec()</b> returns PCRE_ERROR_PARTIAL instead of
1244    PCRE_ERROR_NOMATCH. When PCRE_PARTIAL is used, there are restrictions on what
1245    may appear in the pattern. These are discussed in the
1246    <a href="pcrepartial.html"><b>pcrepartial</b></a>
1247    documentation.
1248  </P>  </P>
1249    <br><b>
1250    The string to be matched by <b>pcre_exec()</b>
1251    </b><br>
1252  <P>  <P>
1253  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
1254  <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
1255  <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
1256  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
1257  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
1258  </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.  
1259  </P>  </P>
1260  <P>  <P>
1261  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 1263  same subject by calling <b>pcre_exec()</
1263  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
1264  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
1265  lookbehind. For example, consider the pattern  lookbehind. For example, consider the pattern
 </P>  
 <P>  
1266  <pre>  <pre>
1267    \Biss\B    \Biss\B
1268  </PRE>  </pre>
 </P>  
 <P>  
1269  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
1270  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
1271  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 1278  behind the starting point to discover th
1278  </P>  </P>
1279  <P>  <P>
1280  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
1281  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
1282  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.
1283  </P>  </P>
1284    <br><b>
1285    How <b>pcre_exec()</b> returns captured substrings
1286    </b><br>
1287  <P>  <P>
1288  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
1289  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 1295  kinds of parenthesized subpattern that d
1295  <P>  <P>
1296  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
1297  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
1298  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>:
1299  back captured substrings, each substring using a pair of integers. The  this argument is NOT the size of <i>ovector</i> in bytes.
1300  remaining third of the vector is used as workspace by <b>pcre_exec()</b> while  </P>
1301  matching capturing subpatterns, and is not available for passing back  <P>
1302  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,
1303  three. If it is not, it is rounded down.  each substring using a pair of integers. The remaining third of the vector is
1304    used as workspace by <b>pcre_exec()</b> while matching capturing subpatterns,
1305    and is not available for passing back information. The length passed in
1306    <i>ovecsize</i> should always be a multiple of three. If it is not, it is
1307    rounded down.
1308  </P>  </P>
1309  <P>  <P>
1310  When a match has been successful, information about captured substrings is  When a match is successful, information about captured substrings is returned
1311  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
1312  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
1313  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
1314  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
1315  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
1316  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
1317  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>
1318  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
1319  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
1320  just the first pair of offsets has been set.  capturing subpatterns, the return value from a successful match is 1,
1321  </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.  
1322  </P>  </P>
1323  <P>  <P>
1324  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
1325  string that it matched that gets returned.  string that it matched that is returned.
1326  </P>  </P>
1327  <P>  <P>
1328  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
1329  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
1330  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
1331  <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
1332  <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
1333  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
1334  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
1335  to supply an <i>ovector</i>.  advisable to supply an <i>ovector</i>.
1336  </P>  </P>
1337  <P>  <P>
1338  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
1339  subpatterns there are in a compiled pattern. The smallest size for  subpatterns there are in a compiled pattern. The smallest size for
1340  <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
1341  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.
1342  </P>  </P>
1343  <P>  <P>
1344  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
1345  defined in the header file:  the subject when subpattern <i>n</i> has not been used at all. For example, if
1346    the string "abc" is matched against the pattern (a|(z))(bc) the return from the
1347    function is 4, and subpatterns 1 and 3 are matched, but 2 is not. When this
1348    happens, both values in the offset pairs corresponding to unused subpatterns
1349    are set to -1.
1350  </P>  </P>
1351  <P>  <P>
1352  <pre>  Offset values that correspond to unused subpatterns at the end of the
1353    PCRE_ERROR_NOMATCH        (-1)  expression are also set to -1. For example, if the string "abc" is matched
1354  </PRE>  against the pattern (abc)(x(yz)?)? subpatterns 2 and 3 are not matched. The
1355    return from the function is 2, because the highest used capturing subpattern
1356    number is 1. However, you can refer to the offsets for the second and third
1357    capturing subpatterns if you wish (assuming the vector is large enough, of
1358    course).
1359  </P>  </P>
1360  <P>  <P>
1361  The subject string did not match the pattern.  Some convenience functions are provided for extracting the captured substrings
1362  </P>  as separate strings. These are described below.
1363    <a name="errorlist"></a></P>
1364    <br><b>
1365    Error return values from <b>pcre_exec()</b>
1366    </b><br>
1367  <P>  <P>
1368    If <b>pcre_exec()</b> fails, it returns a negative number. The following are
1369    defined in the header file:
1370    <pre>
1371      PCRE_ERROR_NOMATCH        (-1)
1372    </pre>
1373    The subject string did not match the pattern.
1374  <pre>  <pre>
1375    PCRE_ERROR_NULL           (-2)    PCRE_ERROR_NULL           (-2)
1376  </PRE>  </pre>
 </P>  
 <P>  
1377  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
1378  NULL and <i>ovecsize</i> was not zero.  NULL and <i>ovecsize</i> was not zero.
 </P>  
 <P>  
1379  <pre>  <pre>
1380    PCRE_ERROR_BADOPTION      (-3)    PCRE_ERROR_BADOPTION      (-3)
1381  </PRE>  </pre>
 </P>  
 <P>  
1382  An unrecognized bit was set in the <i>options</i> argument.  An unrecognized bit was set in the <i>options</i> argument.
 </P>  
 <P>  
1383  <pre>  <pre>
1384    PCRE_ERROR_BADMAGIC       (-4)    PCRE_ERROR_BADMAGIC       (-4)
1385  </PRE>  </pre>
 </P>  
 <P>  
1386  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
1387  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
1388  magic number isn't present.  compiled in an environment of one endianness is run in an environment with the
1389  </P>  other endianness. This is the error that PCRE gives when the magic number is
1390  <P>  not present.
1391  <pre>  <pre>
1392    PCRE_ERROR_UNKNOWN_NODE   (-5)    PCRE_ERROR_UNKNOWN_OPCODE (-5)
1393  </PRE>  </pre>
 </P>  
 <P>  
1394  While running the pattern match, an unknown item was encountered in the  While running the pattern match, an unknown item was encountered in the
1395  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
1396  of the compiled pattern.  of the compiled pattern.
 </P>  
 <P>  
1397  <pre>  <pre>
1398    PCRE_ERROR_NOMEMORY       (-6)    PCRE_ERROR_NOMEMORY       (-6)
1399  </PRE>  </pre>
 </P>  
 <P>  
1400  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
1401  <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
1402  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
1403  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
1404  the end of matching.  automatically freed at the end of matching.
 </P>  
 <P>  
1405  <pre>  <pre>
1406    PCRE_ERROR_NOSUBSTRING    (-7)    PCRE_ERROR_NOSUBSTRING    (-7)
1407  </PRE>  </pre>
 </P>  
 <P>  
1408  This error is used by the <b>pcre_copy_substring()</b>,  This error is used by the <b>pcre_copy_substring()</b>,
1409  <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
1410  below). It is never returned by <b>pcre_exec()</b>.  below). It is never returned by <b>pcre_exec()</b>.
 </P>  
 <P>  
1411  <pre>  <pre>
1412    PCRE_ERROR_MATCHLIMIT     (-8)    PCRE_ERROR_MATCHLIMIT     (-8)
1413  </PRE>  </pre>
1414  </P>  The backtracking limit, as specified by the <i>match_limit</i> field in a
1415  <P>  <b>pcre_extra</b> structure (or defaulted) was reached. See the description
1416  The recursion and backtracking limit, as specified by the <i>match_limit</i>  above.
1417    <pre>
1418      PCRE_ERROR_CALLOUT        (-9)
1419    </pre>
1420    This error is never generated by <b>pcre_exec()</b> itself. It is provided for
1421    use by callout functions that want to yield a distinctive error code. See the
1422    <a href="pcrecallout.html"><b>pcrecallout</b></a>
1423    documentation for details.
1424    <pre>
1425      PCRE_ERROR_BADUTF8        (-10)
1426    </pre>
1427    A string that contains an invalid UTF-8 byte sequence was passed as a subject.
1428    <pre>
1429      PCRE_ERROR_BADUTF8_OFFSET (-11)
1430    </pre>
1431    The UTF-8 byte sequence that was passed as a subject was valid, but the value
1432    of <i>startoffset</i> did not point to the beginning of a UTF-8 character.
1433    <pre>
1434      PCRE_ERROR_PARTIAL        (-12)
1435    </pre>
1436    The subject string did not match, but it did match partially. See the
1437    <a href="pcrepartial.html"><b>pcrepartial</b></a>
1438    documentation for details of partial matching.
1439    <pre>
1440      PCRE_ERROR_BADPARTIAL     (-13)
1441    </pre>
1442    The PCRE_PARTIAL option was used with a compiled pattern containing items that
1443    are not supported for partial matching. See the
1444    <a href="pcrepartial.html"><b>pcrepartial</b></a>
1445    documentation for details of partial matching.
1446    <pre>
1447      PCRE_ERROR_INTERNAL       (-14)
1448    </pre>
1449    An unexpected internal error has occurred. This error could be caused by a bug
1450    in PCRE or by overwriting of the compiled pattern.
1451    <pre>
1452      PCRE_ERROR_BADCOUNT       (-15)
1453    </pre>
1454    This error is given if the value of the <i>ovecsize</i> argument is negative.
1455    <pre>
1456      PCRE_ERROR_RECURSIONLIMIT (-21)
1457    </pre>
1458    The internal recursion limit, as specified by the <i>match_limit_recursion</i>
1459  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
1460  description above.  description above.
 </P>  
 <P>  
1461  <pre>  <pre>
1462    PCRE_ERROR_CALLOUT        (-9)    PCRE_ERROR_NULLWSLIMIT    (-22)
1463  </PRE>  </pre>
1464    When a group that can match an empty substring is repeated with an unbounded
1465    upper limit, the subject position at the start of the group must be remembered,
1466    so that a test for an empty string can be made when the end of the group is
1467    reached. Some workspace is required for this; if it runs out, this error is
1468    given.
1469    <pre>
1470      PCRE_ERROR_BADNEWLINE     (-23)
1471    </pre>
1472    An invalid combination of PCRE_NEWLINE_<i>xxx</i> options was given.
1473  </P>  </P>
1474  <P>  <P>
1475  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.  
1476  </P>  </P>
1477  <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>
1478  <P>  <P>
1479  <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>
1480  <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 1496  Captured substrings can be accessed dire
1496  <b>pcre_get_substring_list()</b> are provided for extracting captured substrings  <b>pcre_get_substring_list()</b> are provided for extracting captured substrings
1497  as new, separate, zero-terminated strings. These functions identify substrings  as new, separate, zero-terminated strings. These functions identify substrings
1498  by number. The next section describes functions for extracting named  by number. The next section describes functions for extracting named
1499  substrings. A substring that contains a binary zero is correctly extracted and  substrings.
1500  has a further zero added on the end, but the result is not, of course,  </P>
1501  a C string.  <P>
1502    A substring that contains a binary zero is correctly extracted and has a
1503    further zero added on the end, but the result is not, of course, a C string.
1504    However, you can process such a string by referring to the length that is
1505    returned by <b>pcre_copy_substring()</b> and <b>pcre_get_substring()</b>.
1506    Unfortunately, the interface to <b>pcre_get_substring_list()</b> is not adequate
1507    for handling strings containing binary zeros, because the end of the final
1508    string is not independently indicated.
1509  </P>  </P>
1510  <P>  <P>
1511  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:
1512  <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,
1513  <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
1514  <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
1515  captured by the match, including the substring that matched the entire regular  captured by the match, including the substring that matched the entire regular
1516  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
1517  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
1518  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
1519  the vector divided by three.  number of elements in the vector divided by three.
1520  </P>  </P>
1521  <P>  <P>
1522  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>
1523  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
1524  value of zero extracts the substring that matched the entire pattern, while  value of zero extracts the substring that matched the entire pattern, whereas
1525  higher values extract the captured substrings. For <b>pcre_copy_substring()</b>,  higher values extract the captured substrings. For <b>pcre_copy_substring()</b>,
1526  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
1527  <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
1528  obtained via <b>pcre_malloc</b>, and its address is returned via  obtained via <b>pcre_malloc</b>, and its address is returned via
1529  <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
1530  including the terminating zero, or one of  including the terminating zero, or one of these error codes:
 </P>  
 <P>  
1531  <pre>  <pre>
1532    PCRE_ERROR_NOMEMORY       (-6)    PCRE_ERROR_NOMEMORY       (-6)
1533  </PRE>  </pre>
 </P>  
 <P>  
1534  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
1535  memory failed for <b>pcre_get_substring()</b>.  memory failed for <b>pcre_get_substring()</b>.
 </P>  
 <P>  
1536  <pre>  <pre>
1537    PCRE_ERROR_NOSUBSTRING    (-7)    PCRE_ERROR_NOSUBSTRING    (-7)
1538  </PRE>  </pre>
 </P>  
 <P>  
1539  There is no substring whose number is <i>stringnumber</i>.  There is no substring whose number is <i>stringnumber</i>.
1540  </P>  </P>
1541  <P>  <P>
1542  The <b>pcre_get_substring_list()</b> function extracts all available substrings  The <b>pcre_get_substring_list()</b> function extracts all available substrings
1543  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
1544  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
1545  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
1546  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
1547  function is zero if all went well, or  function is zero if all went well, or the error code
 </P>  
 <P>  
1548  <pre>  <pre>
1549    PCRE_ERROR_NOMEMORY       (-6)    PCRE_ERROR_NOMEMORY       (-6)
1550  </PRE>  </pre>
 </P>  
 <P>  
1551  if the attempt to get the memory block failed.  if the attempt to get the memory block failed.
1552  </P>  </P>
1553  <P>  <P>
# Line 1202  a previous call of <b>pcre_get_substring Line 1565  a previous call of <b>pcre_get_substring
1565  <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
1566  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
1567  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
1568  linked via a special interface to another programming language which cannot use  linked via a special interface to another programming language that cannot use
1569  <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
1570  provided.  provided.
1571  </P>  </P>
1572  <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>
1573    <P>
1574    <b>int pcre_get_stringnumber(const pcre *<i>code</i>,</b>
1575    <b>const char *<i>name</i>);</b>
1576    </P>
1577  <P>  <P>
1578  <b>int pcre_copy_named_substring(const pcre *<i>code</i>,</b>  <b>int pcre_copy_named_substring(const pcre *<i>code</i>,</b>
1579  <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 1581  provided.
1581  <b>char *<i>buffer</i>, int <i>buffersize</i>);</b>  <b>char *<i>buffer</i>, int <i>buffersize</i>);</b>
1582  </P>  </P>
1583  <P>  <P>
 <b>int pcre_get_stringnumber(const pcre *<i>code</i>,</b>  
 <b>const char *<i>name</i>);</b>  
 </P>  
 <P>  
1584  <b>int pcre_get_named_substring(const pcre *<i>code</i>,</b>  <b>int pcre_get_named_substring(const pcre *<i>code</i>,</b>
1585  <b>const char *<i>subject</i>, int *<i>ovector</i>,</b>  <b>const char *<i>subject</i>, int *<i>ovector</i>,</b>
1586  <b>int <i>stringcount</i>, const char *<i>stringname</i>,</b>  <b>int <i>stringcount</i>, const char *<i>stringname</i>,</b>
1587  <b>const char **<i>stringptr</i>);</b>  <b>const char **<i>stringptr</i>);</b>
1588  </P>  </P>
1589  <P>  <P>
1590  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.
1591  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>  
1592  <pre>  <pre>
1593    ab(?&#60;xxx&#62;\d+)...    (a+)b(?&#60;xxx&#62;\d+)...
1594  </PRE>  </pre>
1595  </P>  the number of the subpattern called "xxx" is 2. If the name is known to be
1596  <P>  unique (PCRE_DUPNAMES was not set), you can find the number from the name by
1597  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
1598  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
1599  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
1600  whole job.  that name.
1601  </P>  </P>
1602  <P>  <P>
1603  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
1604  <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
1605  extract by number, and so are not re-described here. There are just two  two functions that do the whole job.
1606  differences.  </P>
1607    <P>
1608    Most of the arguments of <b>pcre_copy_named_substring()</b> and
1609    <b>pcre_get_named_substring()</b> are the same as those for the similarly named
1610    functions that extract by number. As these are described in the previous
1611    section, they are not re-described here. There are just two differences:
1612  </P>  </P>
1613  <P>  <P>
1614  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 1618  translation table.
1618  </P>  </P>
1619  <P>  <P>
1620  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
1621  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
1622  appropriate.  appropriate. <b>NOTE:</b> If PCRE_DUPNAMES is set and there are duplicate names,
1623    the behaviour may not be what you want (see the next section).
1624    </P>
1625    <br><a name="SEC17" href="#TOC1">DUPLICATE SUBPATTERN NAMES</a><br>
1626    <P>
1627    <b>int pcre_get_stringtable_entries(const pcre *<i>code</i>,</b>
1628    <b>const char *<i>name</i>, char **<i>first</i>, char **<i>last</i>);</b>
1629    </P>
1630    <P>
1631    When a pattern is compiled with the PCRE_DUPNAMES option, names for subpatterns
1632    are not required to be unique. Normally, patterns with duplicate names are such
1633    that in any one match, only one of the named subpatterns participates. An
1634    example is shown in the
1635    <a href="pcrepattern.html"><b>pcrepattern</b></a>
1636    documentation. When duplicates are present, <b>pcre_copy_named_substring()</b>
1637    and <b>pcre_get_named_substring()</b> return the first substring corresponding
1638    to the given name that is set. If none are set, an empty string is returned.
1639    The <b>pcre_get_stringnumber()</b> function returns one of the numbers that are
1640    associated with the name, but it is not defined which it is.
1641    <br>
1642    <br>
1643    If you want to get full details of all captured substrings for a given name,
1644    you must use the <b>pcre_get_stringtable_entries()</b> function. The first
1645    argument is the compiled pattern, and the second is the name. The third and
1646    fourth are pointers to variables which are updated by the function. After it
1647    has run, they point to the first and last entries in the name-to-number table
1648    for the given name. The function itself returns the length of each entry, or
1649    PCRE_ERROR_NOSUBSTRING (-7) if there are none. The format of the table is
1650    described above in the section entitled <i>Information about a pattern</i>.
1651    Given all the relevant entries for the name, you can extract each of their
1652    numbers, and hence the captured data, if any.
1653    </P>
1654    <br><a name="SEC18" href="#TOC1">FINDING ALL POSSIBLE MATCHES</a><br>
1655    <P>
1656    The traditional matching function uses a similar algorithm to Perl, which stops
1657    when it finds the first match, starting at a given point in the subject. If you
1658    want to find all possible matches, or the longest possible match, consider
1659    using the alternative matching function (see below) instead. If you cannot use
1660    the alternative function, but still need to find all possible matches, you
1661    can kludge it up by making use of the callout facility, which is described in
1662    the
1663    <a href="pcrecallout.html"><b>pcrecallout</b></a>
1664    documentation.
1665    </P>
1666    <P>
1667    What you have to do is to insert a callout right at the end of the pattern.
1668    When your callout function is called, extract and save the current matched
1669    substring. Then return 1, which forces <b>pcre_exec()</b> to backtrack and try
1670    other alternatives. Ultimately, when it runs out of matches, <b>pcre_exec()</b>
1671    will yield PCRE_ERROR_NOMATCH.
1672    <a name="dfamatch"></a></P>
1673    <br><a name="SEC19" href="#TOC1">MATCHING A PATTERN: THE ALTERNATIVE FUNCTION</a><br>
1674    <P>
1675    <b>int pcre_dfa_exec(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>
1676    <b>const char *<i>subject</i>, int <i>length</i>, int <i>startoffset</i>,</b>
1677    <b>int <i>options</i>, int *<i>ovector</i>, int <i>ovecsize</i>,</b>
1678    <b>int *<i>workspace</i>, int <i>wscount</i>);</b>
1679    </P>
1680    <P>
1681    The function <b>pcre_dfa_exec()</b> is called to match a subject string against
1682    a compiled pattern, using a matching algorithm that scans the subject string
1683    just once, and does not backtrack. This has different characteristics to the
1684    normal algorithm, and is not compatible with Perl. Some of the features of PCRE
1685    patterns are not supported. Nevertheless, there are times when this kind of
1686    matching can be useful. For a discussion of the two matching algorithms, see
1687    the
1688    <a href="pcrematching.html"><b>pcrematching</b></a>
1689    documentation.
1690    </P>
1691    <P>
1692    The arguments for the <b>pcre_dfa_exec()</b> function are the same as for
1693    <b>pcre_exec()</b>, plus two extras. The <i>ovector</i> argument is used in a
1694    different way, and this is described below. The other common arguments are used
1695    in the same way as for <b>pcre_exec()</b>, so their description is not repeated
1696    here.
1697    </P>
1698    <P>
1699    The two additional arguments provide workspace for the function. The workspace
1700    vector should contain at least 20 elements. It is used for keeping track of
1701    multiple paths through the pattern tree. More workspace will be needed for
1702    patterns and subjects where there are a lot of potential matches.
1703    </P>
1704    <P>
1705    Here is an example of a simple call to <b>pcre_dfa_exec()</b>:
1706    <pre>
1707      int rc;
1708      int ovector[10];
1709      int wspace[20];
1710      rc = pcre_dfa_exec(
1711        re,             /* result of pcre_compile() */
1712        NULL,           /* we didn't study the pattern */
1713        "some string",  /* the subject string */
1714        11,             /* the length of the subject string */
1715        0,              /* start at offset 0 in the subject */
1716        0,              /* default options */
1717        ovector,        /* vector of integers for substring information */
1718        10,             /* number of elements (NOT size in bytes) */
1719        wspace,         /* working space vector */
1720        20);            /* number of elements (NOT size in bytes) */
1721    </PRE>
1722    </P>
1723    <br><b>
1724    Option bits for <b>pcre_dfa_exec()</b>
1725    </b><br>
1726    <P>
1727    The unused bits of the <i>options</i> argument for <b>pcre_dfa_exec()</b> must be
1728    zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_<i>xxx</i>,
1729    PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NO_UTF8_CHECK, PCRE_PARTIAL,
1730    PCRE_DFA_SHORTEST, and PCRE_DFA_RESTART. All but the last three of these are
1731    the same as for <b>pcre_exec()</b>, so their description is not repeated here.
1732    <pre>
1733      PCRE_PARTIAL
1734    </pre>
1735    This has the same general effect as it does for <b>pcre_exec()</b>, but the
1736    details are slightly different. When PCRE_PARTIAL is set for
1737    <b>pcre_dfa_exec()</b>, the return code PCRE_ERROR_NOMATCH is converted into
1738    PCRE_ERROR_PARTIAL if the end of the subject is reached, there have been no
1739    complete matches, but there is still at least one matching possibility. The
1740    portion of the string that provided the partial match is set as the first
1741    matching string.
1742    <pre>
1743      PCRE_DFA_SHORTEST
1744    </pre>
1745    Setting the PCRE_DFA_SHORTEST option causes the matching algorithm to stop as
1746    soon as it has found one match. Because of the way the alternative algorithm
1747    works, this is necessarily the shortest possible match at the first possible
1748    matching point in the subject string.
1749    <pre>
1750      PCRE_DFA_RESTART
1751    </pre>
1752    When <b>pcre_dfa_exec()</b> is called with the PCRE_PARTIAL option, and returns
1753    a partial match, it is possible to call it again, with additional subject
1754    characters, and have it continue with the same match. The PCRE_DFA_RESTART
1755    option requests this action; when it is set, the <i>workspace</i> and
1756    <i>wscount</i> options must reference the same vector as before because data
1757    about the match so far is left in them after a partial match. There is more
1758    discussion of this facility in the
1759    <a href="pcrepartial.html"><b>pcrepartial</b></a>
1760    documentation.
1761    </P>
1762    <br><b>
1763    Successful returns from <b>pcre_dfa_exec()</b>
1764    </b><br>
1765    <P>
1766    When <b>pcre_dfa_exec()</b> succeeds, it may have matched more than one
1767    substring in the subject. Note, however, that all the matches from one run of
1768    the function start at the same point in the subject. The shorter matches are
1769    all initial substrings of the longer matches. For example, if the pattern
1770    <pre>
1771      &#60;.*&#62;
1772    </pre>
1773    is matched against the string
1774    <pre>
1775      This is &#60;something&#62; &#60;something else&#62; &#60;something further&#62; no more
1776    </pre>
1777    the three matched strings are
1778    <pre>
1779      &#60;something&#62;
1780      &#60;something&#62; &#60;something else&#62;
1781      &#60;something&#62; &#60;something else&#62; &#60;something further&#62;
1782    </pre>
1783    On success, the yield of the function is a number greater than zero, which is
1784    the number of matched substrings. The substrings themselves are returned in
1785    <i>ovector</i>. Each string uses two elements; the first is the offset to the
1786    start, and the second is the offset to the end. In fact, all the strings have
1787    the same start offset. (Space could have been saved by giving this only once,
1788    but it was decided to retain some compatibility with the way <b>pcre_exec()</b>
1789    returns data, even though the meaning of the strings is different.)
1790    </P>
1791    <P>
1792    The strings are returned in reverse order of length; that is, the longest
1793    matching string is given first. If there were too many matches to fit into
1794    <i>ovector</i>, the yield of the function is zero, and the vector is filled with
1795    the longest matches.
1796    </P>
1797    <br><b>
1798    Error returns from <b>pcre_dfa_exec()</b>
1799    </b><br>
1800    <P>
1801    The <b>pcre_dfa_exec()</b> function returns a negative number when it fails.
1802    Many of the errors are the same as for <b>pcre_exec()</b>, and these are
1803    described
1804    <a href="#errorlist">above.</a>
1805    There are in addition the following errors that are specific to
1806    <b>pcre_dfa_exec()</b>:
1807    <pre>
1808      PCRE_ERROR_DFA_UITEM      (-16)
1809    </pre>
1810    This return is given if <b>pcre_dfa_exec()</b> encounters an item in the pattern
1811    that it does not support, for instance, the use of \C or a back reference.
1812    <pre>
1813      PCRE_ERROR_DFA_UCOND      (-17)
1814    </pre>
1815    This return is given if <b>pcre_dfa_exec()</b> encounters a condition item that
1816    uses a back reference for the condition, or a test for recursion in a specific
1817    group. These are not supported.
1818    <pre>
1819      PCRE_ERROR_DFA_UMLIMIT    (-18)
1820    </pre>
1821    This return is given if <b>pcre_dfa_exec()</b> is called with an <i>extra</i>
1822    block that contains a setting of the <i>match_limit</i> field. This is not
1823    supported (it is meaningless).
1824    <pre>
1825      PCRE_ERROR_DFA_WSSIZE     (-19)
1826    </pre>
1827    This return is given if <b>pcre_dfa_exec()</b> runs out of space in the
1828    <i>workspace</i> vector.
1829    <pre>
1830      PCRE_ERROR_DFA_RECURSE    (-20)
1831    </pre>
1832    When a recursive subpattern is processed, the matching function calls itself
1833    recursively, using private vectors for <i>ovector</i> and <i>workspace</i>. This
1834    error is given if the output vector is not large enough. This should be
1835    extremely rare, as a vector of size 1000 is used.
1836    </P>
1837    <br><a name="SEC20" href="#TOC1">SEE ALSO</a><br>
1838    <P>
1839    <b>pcrebuild</b>(3), <b>pcrecallout</b>(3), <b>pcrecpp(3)</b>(3),
1840    <b>pcrematching</b>(3), <b>pcrepartial</b>(3), <b>pcreposix</b>(3),
1841    <b>pcreprecompile</b>(3), <b>pcresample</b>(3), <b>pcrestack</b>(3).
1842  </P>  </P>
1843    <br><a name="SEC21" href="#TOC1">AUTHOR</a><br>
1844  <P>  <P>
1845  Last updated: 03 February 2003  Philip Hazel
1846    <br>
1847    University Computing Service
1848    <br>
1849    Cambridge CB2 3QH, England.
1850    <br>
1851    </P>
1852    <br><a name="SEC22" href="#TOC1">REVISION</a><br>
1853    <P>
1854    Last updated: 06 March 2007
1855    <br>
1856    Copyright &copy; 1997-2007 University of Cambridge.
1857  <br>  <br>
1858  Copyright &copy; 1997-2003 University of Cambridge.  <p>
1859    Return to the <a href="index.html">PCRE index page</a>.
1860    </p>

Legend:
Removed from v.65  
changed lines
  Added in v.142

  ViewVC Help
Powered by ViewVC 1.1.5