/[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 128 by ph10, Tue Mar 20 11:46:50 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.
746  </P>  </P>
747  <P>  <P>
748  An alternative set of tables can, however, be supplied. Such tables are built  An internal set of tables is created in the default C locale when PCRE is
749  by calling the <b>pcre_maketables()</b> function, which has no arguments, in the  built. This is used when the final argument of <b>pcre_compile()</b> is NULL,
750  relevant locale. The result can then be passed to <b>pcre_compile()</b> as often  and is sufficient for many applications. An alternative set of tables can,
751  as necessary. For example, to build and use tables that are appropriate for the  however, be supplied. These may be created in a different locale from the
752  French locale (where accented characters with codes greater than 128 are  default. As more and more applications change to using Unicode, the need for
753  treated as letters), the following code could be used:  this locale support is expected to die away.
754  </P>  </P>
755  <P>  <P>
756    External tables are built by calling the <b>pcre_maketables()</b> function,
757    which has no arguments, in the relevant locale. The result can then be passed
758    to <b>pcre_compile()</b> or <b>pcre_exec()</b> as often as necessary. For
759    example, to build and use tables that are appropriate for the French locale
760    (where accented characters with values greater than 128 are treated as letters),
761    the following code could be used:
762  <pre>  <pre>
763    setlocale(LC_CTYPE, "fr");    setlocale(LC_CTYPE, "fr_FR");
764    tables = pcre_maketables();    tables = pcre_maketables();
765    re = pcre_compile(..., tables);    re = pcre_compile(..., tables);
766  </PRE>  </pre>
767    When <b>pcre_maketables()</b> runs, the tables are built in memory that is
768    obtained via <b>pcre_malloc</b>. It is the caller's responsibility to ensure
769    that the memory containing the tables remains available for as long as it is
770    needed.
771  </P>  </P>
772  <P>  <P>
773  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  
774  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>
775  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
776  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
777  in different locales. It is the caller's responsibility to ensure that the  different patterns can be compiled in different locales.
778  memory containing the tables remains available for as long as it is needed.  </P>
779    <P>
780    It is possible to pass a table pointer or NULL (indicating the use of the
781    internal tables) to <b>pcre_exec()</b>. Although not intended for this purpose,
782    this facility could be used to match a pattern in a different locale from the
783    one in which it was compiled. Passing table pointers at run time is discussed
784    below in the section on matching a pattern.
785  </P>  </P>
786  <br><a name="SEC8" href="#TOC1">INFORMATION ABOUT A PATTERN</a><br>  <br><a name="SEC11" href="#TOC1">INFORMATION ABOUT A PATTERN</a><br>
787  <P>  <P>
788  <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>
789  <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 800  the pattern was not studied. The third a
800  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
801  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
802  the following negative numbers:  the following negative numbers:
 </P>  
 <P>  
803  <pre>  <pre>
804    PCRE_ERROR_NULL       the argument <i>code</i> was NULL    PCRE_ERROR_NULL       the argument <i>code</i> was NULL
805                          the argument <i>where</i> was NULL                          the argument <i>where</i> was NULL
806    PCRE_ERROR_BADMAGIC   the "magic number" was not found    PCRE_ERROR_BADMAGIC   the "magic number" was not found
807    PCRE_ERROR_BADOPTION  the value of <i>what</i> was invalid    PCRE_ERROR_BADOPTION  the value of <i>what</i> was invalid
808  </PRE>  </pre>
809  </P>  The "magic number" is placed at the start of each compiled pattern as an simple
810  <P>  check against passing an arbitrary memory pointer. Here is a typical call of
811  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>  
812  <pre>  <pre>
813    int rc;    int rc;
814    unsigned long int length;    size_t length;
815    rc = pcre_fullinfo(    rc = pcre_fullinfo(
816      re,               /* result of pcre_compile() */      re,               /* result of pcre_compile() */
817      pe,               /* result of pcre_study(), or NULL */      pe,               /* result of pcre_study(), or NULL */
818      PCRE_INFO_SIZE,   /* what is required */      PCRE_INFO_SIZE,   /* what is required */
819      &length);         /* where to put the data */      &length);         /* where to put the data */
820  </PRE>  </pre>
 </P>  
 <P>  
821  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
822  as follows:  as follows:
 </P>  
 <P>  
823  <pre>  <pre>
824    PCRE_INFO_BACKREFMAX    PCRE_INFO_BACKREFMAX
825  </PRE>  </pre>
 </P>  
 <P>  
826  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
827  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
828  no back references.  no back references.
 </P>  
 <P>  
829  <pre>  <pre>
830    PCRE_INFO_CAPTURECOUNT    PCRE_INFO_CAPTURECOUNT
831  </PRE>  </pre>
 </P>  
 <P>  
832  Return the number of capturing subpatterns in the pattern. The fourth argument  Return the number of capturing subpatterns in the pattern. The fourth argument
833  should point to an \fbint\fR variable.  should point to an <b>int</b> variable.
834  </P>  <pre>
835  <P>    PCRE_INFO_DEFAULT_TABLES
836    </pre>
837    Return a pointer to the internal default character tables within PCRE. The
838    fourth argument should point to an <b>unsigned char *</b> variable. This
839    information call is provided for internal use by the <b>pcre_study()</b>
840    function. External callers can cause PCRE to use its internal tables by passing
841    a NULL table pointer.
842  <pre>  <pre>
843    PCRE_INFO_FIRSTBYTE    PCRE_INFO_FIRSTBYTE
844  </PRE>  </pre>
 </P>  
 <P>  
845  Return information about the first byte of any matched string, for a  Return information about the first byte of any matched string, for a
846  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>
847  old name is still recognized for backwards compatibility.)  variable. (This option used to be called PCRE_INFO_FIRSTCHAR; the old name is
848  </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  
849  </P>  </P>
850  <P>  <P>
851    If there is a fixed first byte, for example, from a pattern such as
852    (cat|cow|coyote), its value is returned. Otherwise, if either
853    <br>
854    <br>
855  (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
856  starts with "^", or  starts with "^", or
857  </P>  <br>
858  <P>  <br>
859  (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
860  (if it were set, the pattern would be anchored),  (if it were set, the pattern would be anchored),
861  </P>  <br>
862  <P>  <br>
863  -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
864  subject string or after any newline within the string. Otherwise -2 is  subject string or after any newline within the string. Otherwise -2 is
865  returned. For anchored patterns, -2 is returned.  returned. For anchored patterns, -2 is returned.
 </P>  
 <P>  
866  <pre>  <pre>
867    PCRE_INFO_FIRSTTABLE    PCRE_INFO_FIRSTTABLE
868  </PRE>  </pre>
 </P>  
 <P>  
869  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
870  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
871  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
872  fourth argument should point to an <b>unsigned char *</b> variable.  fourth argument should point to an <b>unsigned char *</b> variable.
 </P>  
 <P>  
873  <pre>  <pre>
874    PCRE_INFO_LASTLITERAL    PCRE_INFO_LASTLITERAL
875  </PRE>  </pre>
 </P>  
 <P>  
876  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
877  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
878  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 880  returned. For anchored patterns, a last
880  follows something of variable length. For example, for the pattern  follows something of variable length. For example, for the pattern
881  /^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
882  is -1.  is -1.
 </P>  
 <P>  
883  <pre>  <pre>
884    PCRE_INFO_NAMECOUNT    PCRE_INFO_NAMECOUNT
885    PCRE_INFO_NAMEENTRYSIZE    PCRE_INFO_NAMEENTRYSIZE
886    PCRE_INFO_NAMETABLE    PCRE_INFO_NAMETABLE
887  </PRE>  </pre>
 </P>  
 <P>  
888  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
889  names are just an additional way of identifying the parentheses, which still  names are just an additional way of identifying the parentheses, which still
890  acquire a number. A caller that wants to extract data from a named subpattern  acquire numbers. Several convenience functions such as
891  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
892  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
893  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
894  table for the pattern.  output vector (described with <b>pcre_exec()</b> below). To do the conversion,
895    you need to use the name-to-number map, which is described by these three
896    values.
897  </P>  </P>
898  <P>  <P>
899  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 903  length of the longest name. PCRE_INFO_NA
903  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
904  are the number of the capturing parenthesis, most significant byte first. The  are the number of the capturing parenthesis, most significant byte first. The
905  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
906  alphabetical order. For example, consider the following pattern (assume  alphabetical order. When PCRE_DUPNAMES is set, duplicate names are in order of
907    their parentheses numbers. For example, consider the following pattern (assume
908  PCRE_EXTENDED is set, so white space - including newlines - is ignored):  PCRE_EXTENDED is set, so white space - including newlines - is ignored):
 </P>  
 <P>  
909  <pre>  <pre>
910    (?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) )
911    (?P&#60;month&#62;\d\d) - (?P&#60;day&#62;\d\d) )  </pre>
 </PRE>  
 </P>  
 <P>  
912  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
913  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
914  bytes shows in hex, and undefined bytes shown as ??:  bytes shows in hexadecimal, and undefined bytes shown as ??:
 </P>  
 <P>  
915  <pre>  <pre>
916    00 01 d  a  t  e  00 ??    00 01 d  a  t  e  00 ??
917    00 05 d  a  y  00 ?? ??    00 05 d  a  y  00 ?? ??
918    00 04 m  o  n  t  h  00    00 04 m  o  n  t  h  00
919    00 02 y  e  a  r  00 ??    00 02 y  e  a  r  00 ??
920  </PRE>  </pre>
921  </P>  When writing code to extract data from named subpatterns using the
922  <P>  name-to-number map, remember that the length of the entries is likely to be
923  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>  
924  <pre>  <pre>
925    PCRE_INFO_OPTIONS    PCRE_INFO_OPTIONS
926  </PRE>  </pre>
 </P>  
 <P>  
927  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
928  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
929  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 932  top-level option settings within the pat
932  <P>  <P>
933  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
934  alternatives begin with one of the following:  alternatives begin with one of the following:
 </P>  
 <P>  
935  <pre>  <pre>
936    ^     unless PCRE_MULTILINE is set    ^     unless PCRE_MULTILINE is set
937    \A    always    \A    always
938    \G    always    \G    always
939    .*    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
940            references to the subpattern in which .* appears  </pre>
 </PRE>  
 </P>  
 <P>  
941  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
942  <b>pcre_fullinfo()</b>.  <b>pcre_fullinfo()</b>.
 </P>  
 <P>  
943  <pre>  <pre>
944    PCRE_INFO_SIZE    PCRE_INFO_SIZE
945  </PRE>  </pre>
 </P>  
 <P>  
946  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
947  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
948  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>
949  variable.  variable.
 </P>  
 <P>  
950  <pre>  <pre>
951    PCRE_INFO_STUDYSIZE    PCRE_INFO_STUDYSIZE
952  </PRE>  </pre>
953  </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  
954  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
955  <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
956  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
957  <b>size_t</b> variable.  <b>size_t</b> variable.
958  </P>  </P>
959  <br><a name="SEC9" href="#TOC1">OBSOLETE INFO FUNCTION</a><br>  <br><a name="SEC12" href="#TOC1">OBSOLETE INFO FUNCTION</a><br>
960  <P>  <P>
961  <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>
962  <b>*<i>firstcharptr</i>);</b>  <b>*<i>firstcharptr</i>);</b>
# Line 753  restrictive to return all the available Line 967  restrictive to return all the available
967  programs should use <b>pcre_fullinfo()</b> instead. The yield of  programs should use <b>pcre_fullinfo()</b> instead. The yield of
968  <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
969  following negative numbers:  following negative numbers:
 </P>  
 <P>  
970  <pre>  <pre>
971    PCRE_ERROR_NULL       the argument <i>code</i> was NULL    PCRE_ERROR_NULL       the argument <i>code</i> was NULL
972    PCRE_ERROR_BADMAGIC   the "magic number" was not found    PCRE_ERROR_BADMAGIC   the "magic number" was not found
973  </PRE>  </pre>
 </P>  
 <P>  
974  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
975  pattern was compiled is placed in the integer it points to (see  pattern was compiled is placed in the integer it points to (see
976  PCRE_INFO_OPTIONS above).  PCRE_INFO_OPTIONS above).
# Line 770  If the pattern is not anchored and the < Line 980  If the pattern is not anchored and the <
980  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
981  string (see PCRE_INFO_FIRSTBYTE above).  string (see PCRE_INFO_FIRSTBYTE above).
982  </P>  </P>
983  <br><a name="SEC10" href="#TOC1">MATCHING A PATTERN</a><br>  <br><a name="SEC13" href="#TOC1">REFERENCE COUNTS</a><br>
984    <P>
985    <b>int pcre_refcount(pcre *<i>code</i>, int <i>adjust</i>);</b>
986    </P>
987    <P>
988    The <b>pcre_refcount()</b> function is used to maintain a reference count in the
989    data block that contains a compiled pattern. It is provided for the benefit of
990    applications that operate in an object-oriented manner, where different parts
991    of the application may be using the same compiled pattern, but you want to free
992    the block when they are all done.
993    </P>
994    <P>
995    When a pattern is compiled, the reference count field is initialized to zero.
996    It is changed only by calling this function, whose action is to add the
997    <i>adjust</i> value (which may be positive or negative) to it. The yield of the
998    function is the new value. However, the value of the count is constrained to
999    lie between 0 and 65535, inclusive. If the new value is outside these limits,
1000    it is forced to the appropriate limit value.
1001    </P>
1002    <P>
1003    Except when it is zero, the reference count is not correctly preserved if a
1004    pattern is compiled on one host and then transferred to a host whose byte-order
1005    is different. (This seems a highly unlikely scenario.)
1006    </P>
1007    <br><a name="SEC14" href="#TOC1">MATCHING A PATTERN: THE TRADITIONAL FUNCTION</a><br>
1008  <P>  <P>
1009  <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>
1010  <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 1012  string (see PCRE_INFO_FIRSTBYTE above).
1012  </P>  </P>
1013  <P>  <P>
1014  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
1015  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
1016  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
1017  <i>extra</i> argument.  <i>extra</i> argument. This function is the main matching facility of the
1018    library, and it operates in a Perl-like manner. For specialist use there is
1019    also an alternative matching function, which is described
1020    <a href="#dfamatch">below</a>
1021    in the section about the <b>pcre_dfa_exec()</b> function.
1022  </P>  </P>
1023  <P>  <P>
1024  Here is an example of a simple call to <b>pcre_exec()</b>:  In most applications, the pattern will have been compiled (and optionally
1025    studied) in the same process that calls <b>pcre_exec()</b>. However, it is
1026    possible to save compiled patterns and study data, and then use them later
1027    in different processes, possibly even on different hosts. For a discussion
1028    about this, see the
1029    <a href="pcreprecompile.html"><b>pcreprecompile</b></a>
1030    documentation.
1031  </P>  </P>
1032  <P>  <P>
1033    Here is an example of a simple call to <b>pcre_exec()</b>:
1034  <pre>  <pre>
1035    int rc;    int rc;
1036    int ovector[30];    int ovector[30];
# Line 796  Here is an example of a simple call to < Line 1041  Here is an example of a simple call to <
1041      11,             /* the length of the subject string */      11,             /* the length of the subject string */
1042      0,              /* start at offset 0 in the subject */      0,              /* start at offset 0 in the subject */
1043      0,              /* default options */      0,              /* default options */
1044      ovector,        /* vector for substring information */      ovector,        /* vector of integers for substring information */
1045      30);            /* number of elements in the vector */      30);            /* number of elements (NOT size in bytes) */
1046  </PRE>  <a name="extradata"></a></PRE>
1047  </P>  </P>
1048    <br><b>
1049    Extra data for <b>pcre_exec()</b>
1050    </b><br>
1051  <P>  <P>
1052  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>
1053  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
1054  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
1055  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
1056  </P>  fields (not necessarily in this order):
 <P>  
1057  <pre>  <pre>
1058    unsigned long int <i>flags</i>;    unsigned long int <i>flags</i>;
1059    void *<i>study_data</i>;    void *<i>study_data</i>;
1060    unsigned long int <i>match_limit</i>;    unsigned long int <i>match_limit</i>;
1061      unsigned long int <i>match_limit_recursion</i>;
1062    void *<i>callout_data</i>;    void *<i>callout_data</i>;
1063  </PRE>    const unsigned char *<i>tables</i>;
1064  </P>  </pre>
 <P>  
1065  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
1066  are set. The flag bits are:  are set. The flag bits are:
 </P>  
 <P>  
1067  <pre>  <pre>
1068    PCRE_EXTRA_STUDY_DATA    PCRE_EXTRA_STUDY_DATA
1069    PCRE_EXTRA_MATCH_LIMIT    PCRE_EXTRA_MATCH_LIMIT
1070      PCRE_EXTRA_MATCH_LIMIT_RECURSION
1071    PCRE_EXTRA_CALLOUT_DATA    PCRE_EXTRA_CALLOUT_DATA
1072  </PRE>    PCRE_EXTRA_TABLES
1073  </P>  </pre>
 <P>  
1074  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
1075  <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
1076  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
1077  the block by setting the other fields.  the block by setting the other fields and their corresponding flag bits.
1078  </P>  </P>
1079  <P>  <P>
1080  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
1081  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,
1082  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
1083  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.  
1084  </P>  </P>
1085  <P>  <P>
1086  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
1087  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
1088  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
1089  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
1090  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
1091  exceeded, <b>pcre_exec()</b> returns PCRE_ERROR_MATCHLIMIT.  string.
1092  </P>  </P>
1093  <P>  <P>
1094  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
1095  which is described in the <b>pcrecallout</b> documentation.  default is 10 million, which handles all but the most extreme cases. You can
1096  </P>  override the default by suppling <b>pcre_exec()</b> with a <b>pcre_extra</b>
1097  <P>  block in which <i>match_limit</i> is set, and PCRE_EXTRA_MATCH_LIMIT is set in
1098  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
1099  unused bits must be zero. This limits <b>pcre_exec()</b> to matching at the  PCRE_ERROR_MATCHLIMIT.
1100  first matching position. However, if a pattern was compiled with PCRE_ANCHORED,  </P>
1101  or turned out to be anchored by virtue of its contents, it cannot be made  <P>
1102  unachored at matching time.  The <i>match_limit_recursion</i> field is similar to <i>match_limit</i>, but
1103    instead of limiting the total number of times that <b>match()</b> is called, it
1104    limits the depth of recursion. The recursion depth is a smaller number than the
1105    total number of calls, because not all calls to <b>match()</b> are recursive.
1106    This limit is of use only if it is set smaller than <i>match_limit</i>.
1107    </P>
1108    <P>
1109    Limiting the recursion depth limits the amount of stack that can be used, or,
1110    when PCRE has been compiled to use memory on the heap instead of the stack, the
1111    amount of heap memory that can be used.
1112    </P>
1113    <P>
1114    The default value for <i>match_limit_recursion</i> can be set when PCRE is
1115    built; the default default is the same value as the default for
1116    <i>match_limit</i>. You can override the default by suppling <b>pcre_exec()</b>
1117    with a <b>pcre_extra</b> block in which <i>match_limit_recursion</i> is set, and
1118    PCRE_EXTRA_MATCH_LIMIT_RECURSION is set in the <i>flags</i> field. If the limit
1119    is exceeded, <b>pcre_exec()</b> returns PCRE_ERROR_RECURSIONLIMIT.
1120  </P>  </P>
1121  <P>  <P>
1122  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,
1123    which is described in the
1124    <a href="pcrecallout.html"><b>pcrecallout</b></a>
1125    documentation.
1126  </P>  </P>
1127  <P>  <P>
1128    The <i>tables</i> field is used to pass a character tables pointer to
1129    <b>pcre_exec()</b>; this overrides the value that is stored with the compiled
1130    pattern. A non-NULL value is stored with the compiled pattern only if custom
1131    tables were supplied to <b>pcre_compile()</b> via its <i>tableptr</i> argument.
1132    If NULL is passed to <b>pcre_exec()</b> using this mechanism, it forces PCRE's
1133    internal tables to be used. This facility is helpful when re-using patterns
1134    that have been saved after compiling with an external set of tables, because
1135    the external tables might be at a different address when <b>pcre_exec()</b> is
1136    called. See the
1137    <a href="pcreprecompile.html"><b>pcreprecompile</b></a>
1138    documentation for a discussion of saving compiled patterns for later use.
1139    </P>
1140    <br><b>
1141    Option bits for <b>pcre_exec()</b>
1142    </b><br>
1143    <P>
1144    The unused bits of the <i>options</i> argument for <b>pcre_exec()</b> must be
1145    zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_<i>xxx</i>,
1146    PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NO_UTF8_CHECK and PCRE_PARTIAL.
1147    <pre>
1148      PCRE_ANCHORED
1149    </pre>
1150    The PCRE_ANCHORED option limits <b>pcre_exec()</b> to matching at the first
1151    matching position. If a pattern was compiled with PCRE_ANCHORED, or turned out
1152    to be anchored by virtue of its contents, it cannot be made unachored at
1153    matching time.
1154    <pre>
1155      PCRE_NEWLINE_CR
1156      PCRE_NEWLINE_LF
1157      PCRE_NEWLINE_CRLF
1158      PCRE_NEWLINE_ANY
1159    </pre>
1160    These options override the newline definition that was chosen or defaulted when
1161    the pattern was compiled. For details, see the description of
1162    <b>pcre_compile()</b> above. During matching, the newline choice affects the
1163    behaviour of the dot, circumflex, and dollar metacharacters. It may also alter
1164    the way the match position is advanced after a match failure for an unanchored
1165    pattern. When PCRE_NEWLINE_CRLF or PCRE_NEWLINE_ANY is set, and a match attempt
1166    fails when the current position is at a CRLF sequence, the match position is
1167    advanced by two characters instead of one, in other words, to after the CRLF.
1168  <pre>  <pre>
1169    PCRE_NOTBOL    PCRE_NOTBOL
1170  </PRE>  </pre>
1171  </P>  This option specifies that first character of the subject string is not the
1172  <P>  beginning of a line, so the circumflex metacharacter should not match before
1173  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
1174  circumflex metacharacter should not match before it. Setting this without  never to match. This option affects only the behaviour of the circumflex
1175  PCRE_MULTILINE (at compile time) causes circumflex never to match.  metacharacter. It does not affect \A.
 </P>  
 <P>  
1176  <pre>  <pre>
1177    PCRE_NOTEOL    PCRE_NOTEOL
1178  </PRE>  </pre>
1179  </P>  This option specifies that the end of the subject string is not the end of a
1180  <P>  line, so the dollar metacharacter should not match it nor (except in multiline
1181  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
1182  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
1183  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>  
1184  <pre>  <pre>
1185    PCRE_NOTEMPTY    PCRE_NOTEMPTY
1186  </PRE>  </pre>
 </P>  
 <P>  
1187  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
1188  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
1189  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>  
1190  <pre>  <pre>
1191    a?b?    a?b?
1192  </PRE>  </pre>
 </P>  
 <P>  
1193  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
1194  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
1195  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 1199  Perl has no direct equivalent of PCRE_NO
1199  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
1200  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
1201  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
1202  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
1203  below) and trying an ordinary match again.  starting offset (see below) and trying an ordinary match again. There is some
1204    code that demonstrates how to do this in the <i>pcredemo.c</i> sample program.
1205    <pre>
1206      PCRE_NO_UTF8_CHECK
1207    </pre>
1208    When PCRE_UTF8 is set at compile time, the validity of the subject as a UTF-8
1209    string is automatically checked when <b>pcre_exec()</b> is subsequently called.
1210    The value of <i>startoffset</i> is also checked to ensure that it points to the
1211    start of a UTF-8 character. If an invalid UTF-8 sequence of bytes is found,
1212    <b>pcre_exec()</b> returns the error PCRE_ERROR_BADUTF8. If <i>startoffset</i>
1213    contains an invalid value, PCRE_ERROR_BADUTF8_OFFSET is returned.
1214    </P>
1215    <P>
1216    If you already know that your subject is valid, and you want to skip these
1217    checks for performance reasons, you can set the PCRE_NO_UTF8_CHECK option when
1218    calling <b>pcre_exec()</b>. You might want to do this for the second and
1219    subsequent calls to <b>pcre_exec()</b> if you are making repeated calls to find
1220    all the matches in a single subject string. However, you should be sure that
1221    the value of <i>startoffset</i> points to the start of a UTF-8 character. When
1222    PCRE_NO_UTF8_CHECK is set, the effect of passing an invalid UTF-8 string as a
1223    subject, or a value of <i>startoffset</i> that does not point to the start of a
1224    UTF-8 character, is undefined. Your program may crash.
1225    <pre>
1226      PCRE_PARTIAL
1227    </pre>
1228    This option turns on the partial matching feature. If the subject string fails
1229    to match the pattern, but at some point during the matching process the end of
1230    the subject was reached (that is, the subject partially matches the pattern and
1231    the failure to match occurred only because there were not enough subject
1232    characters), <b>pcre_exec()</b> returns PCRE_ERROR_PARTIAL instead of
1233    PCRE_ERROR_NOMATCH. When PCRE_PARTIAL is used, there are restrictions on what
1234    may appear in the pattern. These are discussed in the
1235    <a href="pcrepartial.html"><b>pcrepartial</b></a>
1236    documentation.
1237  </P>  </P>
1238    <br><b>
1239    The string to be matched by <b>pcre_exec()</b>
1240    </b><br>
1241  <P>  <P>
1242  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
1243  <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
1244  <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
1245  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
1246  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
1247  </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.  
1248  </P>  </P>
1249  <P>  <P>
1250  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 1252  same subject by calling <b>pcre_exec()</
1252  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
1253  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
1254  lookbehind. For example, consider the pattern  lookbehind. For example, consider the pattern
 </P>  
 <P>  
1255  <pre>  <pre>
1256    \Biss\B    \Biss\B
1257  </PRE>  </pre>
 </P>  
 <P>  
1258  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
1259  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
1260  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 1267  behind the starting point to discover th
1267  </P>  </P>
1268  <P>  <P>
1269  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
1270  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
1271  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.
1272  </P>  </P>
1273    <br><b>
1274    How <b>pcre_exec()</b> returns captured substrings
1275    </b><br>
1276  <P>  <P>
1277  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
1278  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 1284  kinds of parenthesized subpattern that d
1284  <P>  <P>
1285  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
1286  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
1287  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>:
1288  back captured substrings, each substring using a pair of integers. The  this argument is NOT the size of <i>ovector</i> in bytes.
1289  remaining third of the vector is used as workspace by <b>pcre_exec()</b> while  </P>
1290  matching capturing subpatterns, and is not available for passing back  <P>
1291  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,
1292  three. If it is not, it is rounded down.  each substring using a pair of integers. The remaining third of the vector is
1293    used as workspace by <b>pcre_exec()</b> while matching capturing subpatterns,
1294    and is not available for passing back information. The length passed in
1295    <i>ovecsize</i> should always be a multiple of three. If it is not, it is
1296    rounded down.
1297  </P>  </P>
1298  <P>  <P>
1299  When a match has been successful, information about captured substrings is  When a match is successful, information about captured substrings is returned
1300  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
1301  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
1302  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
1303  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
1304  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
1305  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
1306  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>
1307  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
1308  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
1309  just the first pair of offsets has been set.  capturing subpatterns, the return value from a successful match is 1,
1310  </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.  
1311  </P>  </P>
1312  <P>  <P>
1313  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
1314  string that it matched that gets returned.  string that it matched that is returned.
1315  </P>  </P>
1316  <P>  <P>
1317  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
1318  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
1319  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
1320  <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
1321  <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
1322  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
1323  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
1324  to supply an <i>ovector</i>.  advisable to supply an <i>ovector</i>.
1325  </P>  </P>
1326  <P>  <P>
1327  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
1328  subpatterns there are in a compiled pattern. The smallest size for  subpatterns there are in a compiled pattern. The smallest size for
1329  <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
1330  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.
1331  </P>  </P>
1332  <P>  <P>
1333  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
1334  defined in the header file:  the subject when subpattern <i>n</i> has not been used at all. For example, if
1335    the string "abc" is matched against the pattern (a|(z))(bc) the return from the
1336    function is 4, and subpatterns 1 and 3 are matched, but 2 is not. When this
1337    happens, both values in the offset pairs corresponding to unused subpatterns
1338    are set to -1.
1339  </P>  </P>
1340  <P>  <P>
1341  <pre>  Offset values that correspond to unused subpatterns at the end of the
1342    PCRE_ERROR_NOMATCH        (-1)  expression are also set to -1. For example, if the string "abc" is matched
1343  </PRE>  against the pattern (abc)(x(yz)?)? subpatterns 2 and 3 are not matched. The
1344    return from the function is 2, because the highest used capturing subpattern
1345    number is 1. However, you can refer to the offsets for the second and third
1346    capturing subpatterns if you wish (assuming the vector is large enough, of
1347    course).
1348  </P>  </P>
1349  <P>  <P>
1350  The subject string did not match the pattern.  Some convenience functions are provided for extracting the captured substrings
1351  </P>  as separate strings. These are described below.
1352    <a name="errorlist"></a></P>
1353    <br><b>
1354    Error return values from <b>pcre_exec()</b>
1355    </b><br>
1356  <P>  <P>
1357    If <b>pcre_exec()</b> fails, it returns a negative number. The following are
1358    defined in the header file:
1359    <pre>
1360      PCRE_ERROR_NOMATCH        (-1)
1361    </pre>
1362    The subject string did not match the pattern.
1363  <pre>  <pre>
1364    PCRE_ERROR_NULL           (-2)    PCRE_ERROR_NULL           (-2)
1365  </PRE>  </pre>
 </P>  
 <P>  
1366  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
1367  NULL and <i>ovecsize</i> was not zero.  NULL and <i>ovecsize</i> was not zero.
 </P>  
 <P>  
1368  <pre>  <pre>
1369    PCRE_ERROR_BADOPTION      (-3)    PCRE_ERROR_BADOPTION      (-3)
1370  </PRE>  </pre>
 </P>  
 <P>  
1371  An unrecognized bit was set in the <i>options</i> argument.  An unrecognized bit was set in the <i>options</i> argument.
 </P>  
 <P>  
1372  <pre>  <pre>
1373    PCRE_ERROR_BADMAGIC       (-4)    PCRE_ERROR_BADMAGIC       (-4)
1374  </PRE>  </pre>
 </P>  
 <P>  
1375  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
1376  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
1377  magic number isn't present.  compiled in an environment of one endianness is run in an environment with the
1378  </P>  other endianness. This is the error that PCRE gives when the magic number is
1379  <P>  not present.
1380  <pre>  <pre>
1381    PCRE_ERROR_UNKNOWN_NODE   (-5)    PCRE_ERROR_UNKNOWN_OPCODE (-5)
1382  </PRE>  </pre>
 </P>  
 <P>  
1383  While running the pattern match, an unknown item was encountered in the  While running the pattern match, an unknown item was encountered in the
1384  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
1385  of the compiled pattern.  of the compiled pattern.
 </P>  
 <P>  
1386  <pre>  <pre>
1387    PCRE_ERROR_NOMEMORY       (-6)    PCRE_ERROR_NOMEMORY       (-6)
1388  </PRE>  </pre>
 </P>  
 <P>  
1389  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
1390  <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
1391  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
1392  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
1393  the end of matching.  automatically freed at the end of matching.
 </P>  
 <P>  
1394  <pre>  <pre>
1395    PCRE_ERROR_NOSUBSTRING    (-7)    PCRE_ERROR_NOSUBSTRING    (-7)
1396  </PRE>  </pre>
 </P>  
 <P>  
1397  This error is used by the <b>pcre_copy_substring()</b>,  This error is used by the <b>pcre_copy_substring()</b>,
1398  <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
1399  below). It is never returned by <b>pcre_exec()</b>.  below). It is never returned by <b>pcre_exec()</b>.
 </P>  
 <P>  
1400  <pre>  <pre>
1401    PCRE_ERROR_MATCHLIMIT     (-8)    PCRE_ERROR_MATCHLIMIT     (-8)
1402  </PRE>  </pre>
1403  </P>  The backtracking limit, as specified by the <i>match_limit</i> field in a
1404  <P>  <b>pcre_extra</b> structure (or defaulted) was reached. See the description
1405  The recursion and backtracking limit, as specified by the <i>match_limit</i>  above.
1406    <pre>
1407      PCRE_ERROR_CALLOUT        (-9)
1408    </pre>
1409    This error is never generated by <b>pcre_exec()</b> itself. It is provided for
1410    use by callout functions that want to yield a distinctive error code. See the
1411    <a href="pcrecallout.html"><b>pcrecallout</b></a>
1412    documentation for details.
1413    <pre>
1414      PCRE_ERROR_BADUTF8        (-10)
1415    </pre>
1416    A string that contains an invalid UTF-8 byte sequence was passed as a subject.
1417    <pre>
1418      PCRE_ERROR_BADUTF8_OFFSET (-11)
1419    </pre>
1420    The UTF-8 byte sequence that was passed as a subject was valid, but the value
1421    of <i>startoffset</i> did not point to the beginning of a UTF-8 character.
1422    <pre>
1423      PCRE_ERROR_PARTIAL        (-12)
1424    </pre>
1425    The subject string did not match, but it did match partially. See the
1426    <a href="pcrepartial.html"><b>pcrepartial</b></a>
1427    documentation for details of partial matching.
1428    <pre>
1429      PCRE_ERROR_BADPARTIAL     (-13)
1430    </pre>
1431    The PCRE_PARTIAL option was used with a compiled pattern containing items that
1432    are not supported for partial matching. See the
1433    <a href="pcrepartial.html"><b>pcrepartial</b></a>
1434    documentation for details of partial matching.
1435    <pre>
1436      PCRE_ERROR_INTERNAL       (-14)
1437    </pre>
1438    An unexpected internal error has occurred. This error could be caused by a bug
1439    in PCRE or by overwriting of the compiled pattern.
1440    <pre>
1441      PCRE_ERROR_BADCOUNT       (-15)
1442    </pre>
1443    This error is given if the value of the <i>ovecsize</i> argument is negative.
1444    <pre>
1445      PCRE_ERROR_RECURSIONLIMIT (-21)
1446    </pre>
1447    The internal recursion limit, as specified by the <i>match_limit_recursion</i>
1448  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
1449  description above.  description above.
 </P>  
 <P>  
1450  <pre>  <pre>
1451    PCRE_ERROR_CALLOUT        (-9)    PCRE_ERROR_NULLWSLIMIT    (-22)
1452  </PRE>  </pre>
1453    When a group that can match an empty substring is repeated with an unbounded
1454    upper limit, the subject position at the start of the group must be remembered,
1455    so that a test for an empty string can be made when the end of the group is
1456    reached. Some workspace is required for this; if it runs out, this error is
1457    given.
1458    <pre>
1459      PCRE_ERROR_BADNEWLINE     (-23)
1460    </pre>
1461    An invalid combination of PCRE_NEWLINE_<i>xxx</i> options was given.
1462  </P>  </P>
1463  <P>  <P>
1464  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.  
1465  </P>  </P>
1466  <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>
1467  <P>  <P>
1468  <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>
1469  <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 1485  Captured substrings can be accessed dire
1485  <b>pcre_get_substring_list()</b> are provided for extracting captured substrings  <b>pcre_get_substring_list()</b> are provided for extracting captured substrings
1486  as new, separate, zero-terminated strings. These functions identify substrings  as new, separate, zero-terminated strings. These functions identify substrings
1487  by number. The next section describes functions for extracting named  by number. The next section describes functions for extracting named
1488  substrings. A substring that contains a binary zero is correctly extracted and  substrings.
1489  has a further zero added on the end, but the result is not, of course,  </P>
1490  a C string.  <P>
1491    A substring that contains a binary zero is correctly extracted and has a
1492    further zero added on the end, but the result is not, of course, a C string.
1493    However, you can process such a string by referring to the length that is
1494    returned by <b>pcre_copy_substring()</b> and <b>pcre_get_substring()</b>.
1495    Unfortunately, the interface to <b>pcre_get_substring_list()</b> is not adequate
1496    for handling strings containing binary zeros, because the end of the final
1497    string is not independently indicated.
1498  </P>  </P>
1499  <P>  <P>
1500  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:
1501  <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,
1502  <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
1503  <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
1504  captured by the match, including the substring that matched the entire regular  captured by the match, including the substring that matched the entire regular
1505  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
1506  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
1507  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
1508  the vector divided by three.  number of elements in the vector divided by three.
1509  </P>  </P>
1510  <P>  <P>
1511  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>
1512  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
1513  value of zero extracts the substring that matched the entire pattern, while  value of zero extracts the substring that matched the entire pattern, whereas
1514  higher values extract the captured substrings. For <b>pcre_copy_substring()</b>,  higher values extract the captured substrings. For <b>pcre_copy_substring()</b>,
1515  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
1516  <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
1517  obtained via <b>pcre_malloc</b>, and its address is returned via  obtained via <b>pcre_malloc</b>, and its address is returned via
1518  <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
1519  including the terminating zero, or one of  including the terminating zero, or one of these error codes:
 </P>  
 <P>  
1520  <pre>  <pre>
1521    PCRE_ERROR_NOMEMORY       (-6)    PCRE_ERROR_NOMEMORY       (-6)
1522  </PRE>  </pre>
 </P>  
 <P>  
1523  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
1524  memory failed for <b>pcre_get_substring()</b>.  memory failed for <b>pcre_get_substring()</b>.
 </P>  
 <P>  
1525  <pre>  <pre>
1526    PCRE_ERROR_NOSUBSTRING    (-7)    PCRE_ERROR_NOSUBSTRING    (-7)
1527  </PRE>  </pre>
 </P>  
 <P>  
1528  There is no substring whose number is <i>stringnumber</i>.  There is no substring whose number is <i>stringnumber</i>.
1529  </P>  </P>
1530  <P>  <P>
1531  The <b>pcre_get_substring_list()</b> function extracts all available substrings  The <b>pcre_get_substring_list()</b> function extracts all available substrings
1532  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
1533  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
1534  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
1535  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
1536  function is zero if all went well, or  function is zero if all went well, or the error code
 </P>  
 <P>  
1537  <pre>  <pre>
1538    PCRE_ERROR_NOMEMORY       (-6)    PCRE_ERROR_NOMEMORY       (-6)
1539  </PRE>  </pre>
 </P>  
 <P>  
1540  if the attempt to get the memory block failed.  if the attempt to get the memory block failed.
1541  </P>  </P>
1542  <P>  <P>
# Line 1202  a previous call of <b>pcre_get_substring Line 1554  a previous call of <b>pcre_get_substring
1554  <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
1555  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
1556  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
1557  linked via a special interface to another programming language which cannot use  linked via a special interface to another programming language that cannot use
1558  <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
1559  provided.  provided.
1560  </P>  </P>
1561  <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>
1562    <P>
1563    <b>int pcre_get_stringnumber(const pcre *<i>code</i>,</b>
1564    <b>const char *<i>name</i>);</b>
1565    </P>
1566  <P>  <P>
1567  <b>int pcre_copy_named_substring(const pcre *<i>code</i>,</b>  <b>int pcre_copy_named_substring(const pcre *<i>code</i>,</b>
1568  <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 1570  provided.
1570  <b>char *<i>buffer</i>, int <i>buffersize</i>);</b>  <b>char *<i>buffer</i>, int <i>buffersize</i>);</b>
1571  </P>  </P>
1572  <P>  <P>
 <b>int pcre_get_stringnumber(const pcre *<i>code</i>,</b>  
 <b>const char *<i>name</i>);</b>  
 </P>  
 <P>  
1573  <b>int pcre_get_named_substring(const pcre *<i>code</i>,</b>  <b>int pcre_get_named_substring(const pcre *<i>code</i>,</b>
1574  <b>const char *<i>subject</i>, int *<i>ovector</i>,</b>  <b>const char *<i>subject</i>, int *<i>ovector</i>,</b>
1575  <b>int <i>stringcount</i>, const char *<i>stringname</i>,</b>  <b>int <i>stringcount</i>, const char *<i>stringname</i>,</b>
1576  <b>const char **<i>stringptr</i>);</b>  <b>const char **<i>stringptr</i>);</b>
1577  </P>  </P>
1578  <P>  <P>
1579  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.
1580  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>  
1581  <pre>  <pre>
1582    ab(?&#60;xxx&#62;\d+)...    (a+)b(?&#60;xxx&#62;\d+)...
1583  </PRE>  </pre>
1584  </P>  the number of the subpattern called "xxx" is 2. If the name is known to be
1585  <P>  unique (PCRE_DUPNAMES was not set), you can find the number from the name by
1586  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
1587  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
1588  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
1589  whole job.  that name.
1590  </P>  </P>
1591  <P>  <P>
1592  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
1593  <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
1594  extract by number, and so are not re-described here. There are just two  two functions that do the whole job.
1595  differences.  </P>
1596    <P>
1597    Most of the arguments of <b>pcre_copy_named_substring()</b> and
1598    <b>pcre_get_named_substring()</b> are the same as those for the similarly named
1599    functions that extract by number. As these are described in the previous
1600    section, they are not re-described here. There are just two differences:
1601  </P>  </P>
1602  <P>  <P>
1603  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 1607  translation table.
1607  </P>  </P>
1608  <P>  <P>
1609  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
1610  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
1611  appropriate.  appropriate. <b>NOTE:</b> If PCRE_DUPNAMES is set and there are duplicate names,
1612    the behaviour may not be what you want (see the next section).
1613    </P>
1614    <br><a name="SEC17" href="#TOC1">DUPLICATE SUBPATTERN NAMES</a><br>
1615    <P>
1616    <b>int pcre_get_stringtable_entries(const pcre *<i>code</i>,</b>
1617    <b>const char *<i>name</i>, char **<i>first</i>, char **<i>last</i>);</b>
1618    </P>
1619    <P>
1620    When a pattern is compiled with the PCRE_DUPNAMES option, names for subpatterns
1621    are not required to be unique. Normally, patterns with duplicate names are such
1622    that in any one match, only one of the named subpatterns participates. An
1623    example is shown in the
1624    <a href="pcrepattern.html"><b>pcrepattern</b></a>
1625    documentation. When duplicates are present, <b>pcre_copy_named_substring()</b>
1626    and <b>pcre_get_named_substring()</b> return the first substring corresponding
1627    to the given name that is set. If none are set, an empty string is returned.
1628    The <b>pcre_get_stringnumber()</b> function returns one of the numbers that are
1629    associated with the name, but it is not defined which it is.
1630    <br>
1631    <br>
1632    If you want to get full details of all captured substrings for a given name,
1633    you must use the <b>pcre_get_stringtable_entries()</b> function. The first
1634    argument is the compiled pattern, and the second is the name. The third and
1635    fourth are pointers to variables which are updated by the function. After it
1636    has run, they point to the first and last entries in the name-to-number table
1637    for the given name. The function itself returns the length of each entry, or
1638    PCRE_ERROR_NOSUBSTRING (-7) if there are none. The format of the table is
1639    described above in the section entitled <i>Information about a pattern</i>.
1640    Given all the relevant entries for the name, you can extract each of their
1641    numbers, and hence the captured data, if any.
1642    </P>
1643    <br><a name="SEC18" href="#TOC1">FINDING ALL POSSIBLE MATCHES</a><br>
1644    <P>
1645    The traditional matching function uses a similar algorithm to Perl, which stops
1646    when it finds the first match, starting at a given point in the subject. If you
1647    want to find all possible matches, or the longest possible match, consider
1648    using the alternative matching function (see below) instead. If you cannot use
1649    the alternative function, but still need to find all possible matches, you
1650    can kludge it up by making use of the callout facility, which is described in
1651    the
1652    <a href="pcrecallout.html"><b>pcrecallout</b></a>
1653    documentation.
1654    </P>
1655    <P>
1656    What you have to do is to insert a callout right at the end of the pattern.
1657    When your callout function is called, extract and save the current matched
1658    substring. Then return 1, which forces <b>pcre_exec()</b> to backtrack and try
1659    other alternatives. Ultimately, when it runs out of matches, <b>pcre_exec()</b>
1660    will yield PCRE_ERROR_NOMATCH.
1661    <a name="dfamatch"></a></P>
1662    <br><a name="SEC19" href="#TOC1">MATCHING A PATTERN: THE ALTERNATIVE FUNCTION</a><br>
1663    <P>
1664    <b>int pcre_dfa_exec(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>
1665    <b>const char *<i>subject</i>, int <i>length</i>, int <i>startoffset</i>,</b>
1666    <b>int <i>options</i>, int *<i>ovector</i>, int <i>ovecsize</i>,</b>
1667    <b>int *<i>workspace</i>, int <i>wscount</i>);</b>
1668    </P>
1669    <P>
1670    The function <b>pcre_dfa_exec()</b> is called to match a subject string against
1671    a compiled pattern, using a matching algorithm that scans the subject string
1672    just once, and does not backtrack. This has different characteristics to the
1673    normal algorithm, and is not compatible with Perl. Some of the features of PCRE
1674    patterns are not supported. Nevertheless, there are times when this kind of
1675    matching can be useful. For a discussion of the two matching algorithms, see
1676    the
1677    <a href="pcrematching.html"><b>pcrematching</b></a>
1678    documentation.
1679    </P>
1680    <P>
1681    The arguments for the <b>pcre_dfa_exec()</b> function are the same as for
1682    <b>pcre_exec()</b>, plus two extras. The <i>ovector</i> argument is used in a
1683    different way, and this is described below. The other common arguments are used
1684    in the same way as for <b>pcre_exec()</b>, so their description is not repeated
1685    here.
1686    </P>
1687    <P>
1688    The two additional arguments provide workspace for the function. The workspace
1689    vector should contain at least 20 elements. It is used for keeping track of
1690    multiple paths through the pattern tree. More workspace will be needed for
1691    patterns and subjects where there are a lot of potential matches.
1692    </P>
1693    <P>
1694    Here is an example of a simple call to <b>pcre_dfa_exec()</b>:
1695    <pre>
1696      int rc;
1697      int ovector[10];
1698      int wspace[20];
1699      rc = pcre_dfa_exec(
1700        re,             /* result of pcre_compile() */
1701        NULL,           /* we didn't study the pattern */
1702        "some string",  /* the subject string */
1703        11,             /* the length of the subject string */
1704        0,              /* start at offset 0 in the subject */
1705        0,              /* default options */
1706        ovector,        /* vector of integers for substring information */
1707        10,             /* number of elements (NOT size in bytes) */
1708        wspace,         /* working space vector */
1709        20);            /* number of elements (NOT size in bytes) */
1710    </PRE>
1711    </P>
1712    <br><b>
1713    Option bits for <b>pcre_dfa_exec()</b>
1714    </b><br>
1715    <P>
1716    The unused bits of the <i>options</i> argument for <b>pcre_dfa_exec()</b> must be
1717    zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_<i>xxx</i>,
1718    PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NO_UTF8_CHECK, PCRE_PARTIAL,
1719    PCRE_DFA_SHORTEST, and PCRE_DFA_RESTART. All but the last three of these are
1720    the same as for <b>pcre_exec()</b>, so their description is not repeated here.
1721    <pre>
1722      PCRE_PARTIAL
1723    </pre>
1724    This has the same general effect as it does for <b>pcre_exec()</b>, but the
1725    details are slightly different. When PCRE_PARTIAL is set for
1726    <b>pcre_dfa_exec()</b>, the return code PCRE_ERROR_NOMATCH is converted into
1727    PCRE_ERROR_PARTIAL if the end of the subject is reached, there have been no
1728    complete matches, but there is still at least one matching possibility. The
1729    portion of the string that provided the partial match is set as the first
1730    matching string.
1731    <pre>
1732      PCRE_DFA_SHORTEST
1733    </pre>
1734    Setting the PCRE_DFA_SHORTEST option causes the matching algorithm to stop as
1735    soon as it has found one match. Because of the way the alternative algorithm
1736    works, this is necessarily the shortest possible match at the first possible
1737    matching point in the subject string.
1738    <pre>
1739      PCRE_DFA_RESTART
1740    </pre>
1741    When <b>pcre_dfa_exec()</b> is called with the PCRE_PARTIAL option, and returns
1742    a partial match, it is possible to call it again, with additional subject
1743    characters, and have it continue with the same match. The PCRE_DFA_RESTART
1744    option requests this action; when it is set, the <i>workspace</i> and
1745    <i>wscount</i> options must reference the same vector as before because data
1746    about the match so far is left in them after a partial match. There is more
1747    discussion of this facility in the
1748    <a href="pcrepartial.html"><b>pcrepartial</b></a>
1749    documentation.
1750    </P>
1751    <br><b>
1752    Successful returns from <b>pcre_dfa_exec()</b>
1753    </b><br>
1754    <P>
1755    When <b>pcre_dfa_exec()</b> succeeds, it may have matched more than one
1756    substring in the subject. Note, however, that all the matches from one run of
1757    the function start at the same point in the subject. The shorter matches are
1758    all initial substrings of the longer matches. For example, if the pattern
1759    <pre>
1760      &#60;.*&#62;
1761    </pre>
1762    is matched against the string
1763    <pre>
1764      This is &#60;something&#62; &#60;something else&#62; &#60;something further&#62; no more
1765    </pre>
1766    the three matched strings are
1767    <pre>
1768      &#60;something&#62;
1769      &#60;something&#62; &#60;something else&#62;
1770      &#60;something&#62; &#60;something else&#62; &#60;something further&#62;
1771    </pre>
1772    On success, the yield of the function is a number greater than zero, which is
1773    the number of matched substrings. The substrings themselves are returned in
1774    <i>ovector</i>. Each string uses two elements; the first is the offset to the
1775    start, and the second is the offset to the end. In fact, all the strings have
1776    the same start offset. (Space could have been saved by giving this only once,
1777    but it was decided to retain some compatibility with the way <b>pcre_exec()</b>
1778    returns data, even though the meaning of the strings is different.)
1779    </P>
1780    <P>
1781    The strings are returned in reverse order of length; that is, the longest
1782    matching string is given first. If there were too many matches to fit into
1783    <i>ovector</i>, the yield of the function is zero, and the vector is filled with
1784    the longest matches.
1785    </P>
1786    <br><b>
1787    Error returns from <b>pcre_dfa_exec()</b>
1788    </b><br>
1789    <P>
1790    The <b>pcre_dfa_exec()</b> function returns a negative number when it fails.
1791    Many of the errors are the same as for <b>pcre_exec()</b>, and these are
1792    described
1793    <a href="#errorlist">above.</a>
1794    There are in addition the following errors that are specific to
1795    <b>pcre_dfa_exec()</b>:
1796    <pre>
1797      PCRE_ERROR_DFA_UITEM      (-16)
1798    </pre>
1799    This return is given if <b>pcre_dfa_exec()</b> encounters an item in the pattern
1800    that it does not support, for instance, the use of \C or a back reference.
1801    <pre>
1802      PCRE_ERROR_DFA_UCOND      (-17)
1803    </pre>
1804    This return is given if <b>pcre_dfa_exec()</b> encounters a condition item that
1805    uses a back reference for the condition, or a test for recursion in a specific
1806    group. These are not supported.
1807    <pre>
1808      PCRE_ERROR_DFA_UMLIMIT    (-18)
1809    </pre>
1810    This return is given if <b>pcre_dfa_exec()</b> is called with an <i>extra</i>
1811    block that contains a setting of the <i>match_limit</i> field. This is not
1812    supported (it is meaningless).
1813    <pre>
1814      PCRE_ERROR_DFA_WSSIZE     (-19)
1815    </pre>
1816    This return is given if <b>pcre_dfa_exec()</b> runs out of space in the
1817    <i>workspace</i> vector.
1818    <pre>
1819      PCRE_ERROR_DFA_RECURSE    (-20)
1820    </pre>
1821    When a recursive subpattern is processed, the matching function calls itself
1822    recursively, using private vectors for <i>ovector</i> and <i>workspace</i>. This
1823    error is given if the output vector is not large enough. This should be
1824    extremely rare, as a vector of size 1000 is used.
1825    </P>
1826    <br><a name="SEC20" href="#TOC1">SEE ALSO</a><br>
1827    <P>
1828    <b>pcrebuild</b>(3), <b>pcrecallout</b>(3), <b>pcrecpp(3)</b>(3),
1829    <b>pcrematching</b>(3), <b>pcrepartial</b>(3), <b>pcreposix</b>(3),
1830    <b>pcreprecompile</b>(3), <b>pcresample</b>(3), <b>pcrestack</b>(3).
1831  </P>  </P>
1832    <br><a name="SEC21" href="#TOC1">AUTHOR</a><br>
1833  <P>  <P>
1834  Last updated: 03 February 2003  Philip Hazel
1835    <br>
1836    University Computing Service
1837    <br>
1838    Cambridge CB2 3QH, England.
1839    <br>
1840    </P>
1841    <br><a name="SEC22" href="#TOC1">REVISION</a><br>
1842    <P>
1843    Last updated: 06 March 2007
1844    <br>
1845    Copyright &copy; 1997-2007 University of Cambridge.
1846  <br>  <br>
1847  Copyright &copy; 1997-2003 University of Cambridge.  <p>
1848    Return to the <a href="index.html">PCRE index page</a>.
1849    </p>

Legend:
Removed from v.65  
changed lines
  Added in v.128

  ViewVC Help
Powered by ViewVC 1.1.5