/[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 96 by nigel, Fri Mar 2 13:10:43 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  </ul>  </ul>
37  <br><a name="SEC1" href="#TOC1">SYNOPSIS OF PCRE API</a><br>  <br><a name="SEC1" href="#TOC1">PCRE NATIVE API</a><br>
38  <P>  <P>
39  <b>#include &#60;pcre.h&#62;</b>  <b>#include &#60;pcre.h&#62;</b>
40  </P>  </P>
# Line 30  conversion went wrong.<br> Line 44  conversion went wrong.<br>
44  <b>const unsigned char *<i>tableptr</i>);</b>  <b>const unsigned char *<i>tableptr</i>);</b>
45  </P>  </P>
46  <P>  <P>
47    <b>pcre *pcre_compile2(const char *<i>pattern</i>, int <i>options</i>,</b>
48    <b>int *<i>errorcodeptr</i>,</b>
49    <b>const char **<i>errptr</i>, int *<i>erroffset</i>,</b>
50    <b>const unsigned char *<i>tableptr</i>);</b>
51    </P>
52    <P>
53  <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>
54  <b>const char **<i>errptr</i>);</b>  <b>const char **<i>errptr</i>);</b>
55  </P>  </P>
# Line 39  conversion went wrong.<br> Line 59  conversion went wrong.<br>
59  <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>
60  </P>  </P>
61  <P>  <P>
62    <b>int pcre_dfa_exec(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>
63    <b>const char *<i>subject</i>, int <i>length</i>, int <i>startoffset</i>,</b>
64    <b>int <i>options</i>, int *<i>ovector</i>, int <i>ovecsize</i>,</b>
65    <b>int *<i>workspace</i>, int <i>wscount</i>);</b>
66    </P>
67    <P>
68  <b>int pcre_copy_named_substring(const pcre *<i>code</i>,</b>  <b>int pcre_copy_named_substring(const pcre *<i>code</i>,</b>
69  <b>const char *<i>subject</i>, int *<i>ovector</i>,</b>  <b>const char *<i>subject</i>, int *<i>ovector</i>,</b>
70  <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 86  conversion went wrong.<br>
86  <b>const char *<i>name</i>);</b>  <b>const char *<i>name</i>);</b>
87  </P>  </P>
88  <P>  <P>
89    <b>int pcre_get_stringtable_entries(const pcre *<i>code</i>,</b>
90    <b>const char *<i>name</i>, char **<i>first</i>, char **<i>last</i>);</b>
91    </P>
92    <P>
93  <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>
94  <b>int <i>stringcount</i>, int <i>stringnumber</i>,</b>  <b>int <i>stringcount</i>, int <i>stringnumber</i>,</b>
95  <b>const char **<i>stringptr</i>);</b>  <b>const char **<i>stringptr</i>);</b>
# Line 86  conversion went wrong.<br> Line 116  conversion went wrong.<br>
116  <b>*<i>firstcharptr</i>);</b>  <b>*<i>firstcharptr</i>);</b>
117  </P>  </P>
118  <P>  <P>
119    <b>int pcre_refcount(pcre *<i>code</i>, int <i>adjust</i>);</b>
120    </P>
121    <P>
122  <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>
123  </P>  </P>
124  <P>  <P>
# Line 98  conversion went wrong.<br> Line 131  conversion went wrong.<br>
131  <b>void (*pcre_free)(void *);</b>  <b>void (*pcre_free)(void *);</b>
132  </P>  </P>
133  <P>  <P>
134  <b>int (*pcre_callout)(pcre_callout_block *);</b>  <b>void *(*pcre_stack_malloc)(size_t);</b>
135  </P>  </P>
 <br><a name="SEC2" href="#TOC1">PCRE API</a><br>  
136  <P>  <P>
137  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.  
138  </P>  </P>
139  <P>  <P>
140  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.  
141  </P>  </P>
142    <br><a name="SEC2" href="#TOC1">PCRE API OVERVIEW</a><br>
143  <P>  <P>
144  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
145  are used for compiling and matching regular expressions. A sample program that  also some wrapper functions that correspond to the POSIX regular expression
146  demonstrates the simplest way of using them is given in the file  API. These are described in the
147  <i>pcredemo.c</i>. The <b>pcresample</b> documentation describes how to run it.  <a href="pcreposix.html"><b>pcreposix</b></a>
148    documentation. Both of these APIs define a set of C function calls. A C++
149    wrapper is distributed with PCRE. It is documented in the
150    <a href="pcrecpp.html"><b>pcrecpp</b></a>
151    page.
152  </P>  </P>
153  <P>  <P>
154  There are convenience functions for extracting captured substrings from a  The native API C function prototypes are defined in the header file
155  matched subject string. They are:  <b>pcre.h</b>, and on Unix systems the library itself is called <b>libpcre</b>.
156    It can normally be accessed by adding <b>-lpcre</b> to the command for linking
157    an application that uses PCRE. The header file defines the macros PCRE_MAJOR
158    and PCRE_MINOR to contain the major and minor release numbers for the library.
159    Applications can use these to include support for different releases of PCRE.
160    </P>
161    <P>
162    The functions <b>pcre_compile()</b>, <b>pcre_compile2()</b>, <b>pcre_study()</b>,
163    and <b>pcre_exec()</b> are used for compiling and matching regular expressions
164    in a Perl-compatible manner. A sample program that demonstrates the simplest
165    way of using them is provided in the file called <i>pcredemo.c</i> in the source
166    distribution. The
167    <a href="pcresample.html"><b>pcresample</b></a>
168    documentation describes how to run it.
169    </P>
170    <P>
171    A second matching function, <b>pcre_dfa_exec()</b>, which is not
172    Perl-compatible, is also provided. This uses a different algorithm for the
173    matching. The alternative algorithm finds all possible matches (at a given
174    point in the subject), and scans the subject just once. However, this algorithm
175    does not return captured substrings. A description of the two matching
176    algorithms and their advantages and disadvantages is given in the
177    <a href="pcrematching.html"><b>pcrematching</b></a>
178    documentation.
179  </P>  </P>
180  <P>  <P>
181    In addition to the main compiling and matching functions, there are convenience
182    functions for extracting captured substrings from a subject string that is
183    matched by <b>pcre_exec()</b>. They are:
184  <pre>  <pre>
185    <b>pcre_copy_substring()</b>    <b>pcre_copy_substring()</b>
186    <b>pcre_copy_named_substring()</b>    <b>pcre_copy_named_substring()</b>
187    <b>pcre_get_substring()</b>    <b>pcre_get_substring()</b>
188    <b>pcre_get_named_substring()</b>    <b>pcre_get_named_substring()</b>
189    <b>pcre_get_substring_list()</b>    <b>pcre_get_substring_list()</b>
190  </PRE>    <b>pcre_get_stringnumber()</b>
191  </P>    <b>pcre_get_stringtable_entries()</b>
192  <P>  </pre>
193  <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
194  provided, to free the memory used for extracted strings.  provided, to free the memory used for extracted strings.
195  </P>  </P>
196  <P>  <P>
197  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
198  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>,
199    or <b>pcre_dfa_exec()</b>. This is an optional facility that is provided for
200    specialist use. Most commonly, no special tables are passed, in which case
201    internal tables that are generated when PCRE is built are used.
202  </P>  </P>
203  <P>  <P>
204  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
205  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
206  some of the available information, but is retained for backwards compatibility.  some of the available information, but is retained for backwards compatibility.
207  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
208  version of PCRE and its date of release.  version of PCRE and its date of release.
209  </P>  </P>
210  <P>  <P>
211    The function <b>pcre_refcount()</b> maintains a reference count in a data block
212    containing a compiled pattern. This is provided for the benefit of
213    object-oriented applications.
214    </P>
215    <P>
216  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
217  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,
218  respectively. PCRE calls the memory management functions via these variables,  respectively. PCRE calls the memory management functions via these variables,
219  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
220  should be done before calling any PCRE functions.  should be done before calling any PCRE functions.
221  </P>  </P>
222  <P>  <P>
223    The global variables <b>pcre_stack_malloc</b> and <b>pcre_stack_free</b> are also
224    indirections to memory management functions. These special functions are used
225    only when PCRE is compiled to use the heap for remembering data, instead of
226    recursive function calls, when running the <b>pcre_exec()</b> function. See the
227    <a href="pcrebuild.html"><b>pcrebuild</b></a>
228    documentation for details of how to do this. It is a non-standard way of
229    building PCRE, for use in environments that have limited stacks. Because of the
230    greater use of memory management, it runs more slowly. Separate functions are
231    provided so that special-purpose external code can be used for this case. When
232    used, these functions are always called in a stack-like manner (last obtained,
233    first freed), and always for memory blocks of the same size. There is a
234    discussion about PCRE's stack usage in the
235    <a href="pcrestack.html"><b>pcrestack</b></a>
236    documentation.
237    </P>
238    <P>
239  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
240  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
241  points during a matching operation. Details are given in the <b>pcrecallout</b>  points during a matching operation. Details are given in the
242    <a href="pcrecallout.html"><b>pcrecallout</b></a>
243  documentation.  documentation.
244  </P>  </P>
245  <br><a name="SEC3" href="#TOC1">MULTITHREADING</a><br>  <br><a name="SEC3" href="#TOC1">NEWLINES</a><br>
246    <P>
247    PCRE supports four different conventions for indicating line breaks in
248    strings: a single CR (carriage return) character, a single LF (linefeed)
249    character, the two-character sequence CRLF, or any Unicode newline sequence.
250    The Unicode newline sequences are the three just mentioned, plus the single
251    characters VT (vertical tab, U+000B), FF (formfeed, U+000C), NEL (next line,
252    U+0085), LS (line separator, U+2028), and PS (paragraph separator, U+2029).
253    </P>
254    <P>
255    Each of the first three conventions is used by at least one operating system as
256    its standard newline sequence. When PCRE is built, a default can be specified.
257    The default default is LF, which is the Unix standard. When PCRE is run, the
258    default can be overridden, either when a pattern is compiled, or when it is
259    matched.
260    </P>
261    <P>
262    In the PCRE documentation the word "newline" is used to mean "the character or
263    pair of characters that indicate a line break". The choice of newline
264    convention affects the handling of the dot, circumflex, and dollar
265    metacharacters, the handling of #-comments in /x mode, and, when CRLF is a
266    recognized line ending sequence, the match position advancement for a
267    non-anchored pattern. The choice of newline convention does not affect the
268    interpretation of the \n or \r escape sequences.
269    </P>
270    <br><a name="SEC4" href="#TOC1">MULTITHREADING</a><br>
271  <P>  <P>
272  The PCRE functions can be used in multi-threading applications, with the  The PCRE functions can be used in multi-threading applications, with the
273  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>,
274  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
275  are shared by all threads.  callout function pointed to by <b>pcre_callout</b>, are shared by all threads.
276  </P>  </P>
277  <P>  <P>
278  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
279  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.
280  </P>  </P>
281  <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>
282    <P>
283    The compiled form of a regular expression can be saved and re-used at a later
284    time, possibly by a different program, and even on a host other than the one on
285    which it was compiled. Details are given in the
286    <a href="pcreprecompile.html"><b>pcreprecompile</b></a>
287    documentation.
288    </P>
289    <br><a name="SEC6" href="#TOC1">CHECKING BUILD-TIME OPTIONS</a><br>
290  <P>  <P>
291  <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>
292  </P>  </P>
# Line 186  documentation has more details about the Line 300  documentation has more details about the
300  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
301  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
302  which the information is placed. The following information is available:  which the information is placed. The following information is available:
 </P>  
 <P>  
303  <pre>  <pre>
304    PCRE_CONFIG_UTF8    PCRE_CONFIG_UTF8
305  </PRE>  </pre>
 </P>  
 <P>  
306  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;
307  otherwise it is set to zero.  otherwise it is set to zero.
308  </P>  <pre>
309  <P>    PCRE_CONFIG_UNICODE_PROPERTIES
310    </pre>
311    The output is an integer that is set to one if support for Unicode character
312    properties is available; otherwise it is set to zero.
313  <pre>  <pre>
314    PCRE_CONFIG_NEWLINE    PCRE_CONFIG_NEWLINE
315  </PRE>  </pre>
316  </P>  The output is an integer whose value specifies the default character sequence
317  <P>  that is recognized as meaning "newline". The four values that are supported
318  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
319  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>  
320  <pre>  <pre>
321    PCRE_CONFIG_LINK_SIZE    PCRE_CONFIG_LINK_SIZE
322  </PRE>  </pre>
 </P>  
 <P>  
323  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
324  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
325  allow larger regular expressions to be compiled, at the expense of slower  allow larger regular expressions to be compiled, at the expense of slower
326  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
327  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>  
328  <pre>  <pre>
329    PCRE_CONFIG_POSIX_MALLOC_THRESHOLD    PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
330  </PRE>  </pre>
 </P>  
 <P>  
331  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
332  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
333  the <b>pcreposix</b> documentation.  the
334  </P>  <a href="pcreposix.html"><b>pcreposix</b></a>
335  <P>  documentation.
336  <pre>  <pre>
337    PCRE_CONFIG_MATCH_LIMIT    PCRE_CONFIG_MATCH_LIMIT
338  </PRE>  </pre>
 </P>  
 <P>  
339  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
340  internal matching function calls in a <b>pcre_exec()</b> execution. Further  internal matching function calls in a <b>pcre_exec()</b> execution. Further
341  details are given with <b>pcre_exec()</b> below.  details are given with <b>pcre_exec()</b> below.
342    <pre>
343      PCRE_CONFIG_MATCH_LIMIT_RECURSION
344    </pre>
345    The output is an integer that gives the default limit for the depth of
346    recursion when calling the internal matching function in a <b>pcre_exec()</b>
347    execution. Further details are given with <b>pcre_exec()</b> below.
348    <pre>
349      PCRE_CONFIG_STACKRECURSE
350    </pre>
351    The output is an integer that is set to one if internal recursion when running
352    <b>pcre_exec()</b> is implemented by recursive function calls that use the stack
353    to remember their state. This is the usual way that PCRE is compiled. The
354    output is zero if PCRE was compiled to use blocks of data on the heap instead
355    of recursive function calls. In this case, <b>pcre_stack_malloc</b> and
356    <b>pcre_stack_free</b> are called to manage memory blocks on the heap, thus
357    avoiding the use of the stack.
358  </P>  </P>
359  <br><a name="SEC5" href="#TOC1">COMPILING A PATTERN</a><br>  <br><a name="SEC7" href="#TOC1">COMPILING A PATTERN</a><br>
360  <P>  <P>
361  <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>
362  <b>const char **<i>errptr</i>, int *<i>erroffset</i>,</b>  <b>const char **<i>errptr</i>, int *<i>erroffset</i>,</b>
363  <b>const unsigned char *<i>tableptr</i>);</b>  <b>const unsigned char *<i>tableptr</i>);</b>
364    <b>pcre *pcre_compile2(const char *<i>pattern</i>, int <i>options</i>,</b>
365    <b>int *<i>errorcodeptr</i>,</b>
366    <b>const char **<i>errptr</i>, int *<i>erroffset</i>,</b>
367    <b>const unsigned char *<i>tableptr</i>);</b>
368    </P>
369    <P>
370    Either of the functions <b>pcre_compile()</b> or <b>pcre_compile2()</b> can be
371    called to compile a pattern into an internal form. The only difference between
372    the two interfaces is that <b>pcre_compile2()</b> has an additional argument,
373    <i>errorcodeptr</i>, via which a numerical error code can be returned.
374  </P>  </P>
375  <P>  <P>
376  The function <b>pcre_compile()</b> is called to compile a pattern into an  The pattern is a C string terminated by a binary zero, and is passed in the
377  internal form. The pattern is a C string terminated by a binary zero, and  <i>pattern</i> argument. A pointer to a single block of memory that is obtained
378  is passed in the argument <i>pattern</i>. A pointer to a single block of memory  via <b>pcre_malloc</b> is returned. This contains the compiled code and related
379  that is obtained via <b>pcre_malloc</b> is returned. This contains the compiled  data. The <b>pcre</b> type is defined for the returned block; this is a typedef
380  code and related data. The <b>pcre</b> type is defined for the returned block;  for a structure whose contents are not externally defined. It is up to the
381  this is a typedef for a structure whose contents are not externally defined. It  caller to free the memory (via <b>pcre_free</b>) when it is no longer required.
 is up to the caller to free the memory when it is no longer required.  
382  </P>  </P>
383  <P>  <P>
384  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
385  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
386  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>
387  which is an address (see below).  argument, which is an address (see below).
388  </P>  </P>
389  <P>  <P>
390  The <i>options</i> argument contains independent bits that affect the  The <i>options</i> argument contains various bit settings that affect the
391  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
392  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
393  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
394  in the <b>pcrepattern</b> documentation). For these options, the contents of the  the detailed description in the
395  <i>options</i> argument specifies their initial settings at the start of  <a href="pcrepattern.html"><b>pcrepattern</b></a>
396  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
397    specifies their initial settings at the start of compilation and execution. The
398    PCRE_ANCHORED and PCRE_NEWLINE_<i>xxx</i> options can be set at the time of
399  matching as well as at compile time.  matching as well as at compile time.
400  </P>  </P>
401  <P>  <P>
402  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.
403  Otherwise, if compilation of a pattern fails, <b>pcre_compile()</b> returns  Otherwise, if compilation of a pattern fails, <b>pcre_compile()</b> returns
404  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
405  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
406  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
407    where the error was discovered is placed in the variable pointed to by
408  <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.
409  </P>  </P>
410  <P>  <P>
411  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
412  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
413  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
414  <b>pcre_maketables()</b>. See the section on locale support below.  textual error message. Error codes and messages are listed below.
415  </P>  </P>
416  <P>  <P>
417  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
418    character tables that are built when PCRE is compiled, using the default C
419    locale. Otherwise, <i>tableptr</i> must be an address that is the result of a
420    call to <b>pcre_maketables()</b>. This value is stored with the compiled
421    pattern, and used again by <b>pcre_exec()</b>, unless another table pointer is
422    passed to it. For more discussion, see the section on locale support below.
423  </P>  </P>
424  <P>  <P>
425    This code fragment shows a typical straightforward call to <b>pcre_compile()</b>:
426  <pre>  <pre>
427    pcre *re;    pcre *re;
428    const char *error;    const char *error;
# Line 297  This code fragment shows a typical strai Line 433  This code fragment shows a typical strai
433      &error,           /* for error message */      &error,           /* for error message */
434      &erroffset,       /* for error offset */      &erroffset,       /* for error offset */
435      NULL);            /* use default character tables */      NULL);            /* use default character tables */
436  </PRE>  </pre>
437  </P>  The following names for option bits are defined in the <b>pcre.h</b> header
438  <P>  file:
 The following option bits are defined:  
 </P>  
 <P>  
439  <pre>  <pre>
440    PCRE_ANCHORED    PCRE_ANCHORED
441  </PRE>  </pre>
 </P>  
 <P>  
442  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
443  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
444  being searched (the "subject string"). This effect can also be achieved by  being searched (the "subject string"). This effect can also be achieved by
445  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
446  Perl.  Perl.
447  </P>  <pre>
448  <P>    PCRE_AUTO_CALLOUT
449    </pre>
450    If this bit is set, <b>pcre_compile()</b> automatically inserts callout items,
451    all with number 255, before each pattern item. For discussion of the callout
452    facility, see the
453    <a href="pcrecallout.html"><b>pcrecallout</b></a>
454    documentation.
455  <pre>  <pre>
456    PCRE_CASELESS    PCRE_CASELESS
457  </PRE>  </pre>
 </P>  
 <P>  
458  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
459  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
460  pattern by a (?i) option setting.  pattern by a (?i) option setting. In UTF-8 mode, PCRE always understands the
461  </P>  concept of case for characters whose values are less than 128, so caseless
462  <P>  matching is always possible. For characters with higher values, the concept of
463    case is supported if PCRE is compiled with Unicode property support, but not
464    otherwise. If you want to use caseless matching for characters 128 and above,
465    you must ensure that PCRE is compiled with Unicode property support as well as
466    with UTF-8 support.
467  <pre>  <pre>
468    PCRE_DOLLAR_ENDONLY    PCRE_DOLLAR_ENDONLY
469  </PRE>  </pre>
 </P>  
 <P>  
470  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
471  end of the subject string. Without this option, a dollar also matches  end of the subject string. Without this option, a dollar also matches
472  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
473  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.
474  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
475  a pattern.  pattern.
 </P>  
 <P>  
476  <pre>  <pre>
477    PCRE_DOTALL    PCRE_DOTALL
478  </PRE>  </pre>
 </P>  
 <P>  
479  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,
480  including newlines. Without it, newlines are excluded. This option is  including those that indicate newline. Without it, a dot does not match when
481  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
482  (?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
483  character, independent of the setting of this option.  negative class such as [^a] always matches newline characters, independent of
484  </P>  the setting of this option.
485  <P>  <pre>
486      PCRE_DUPNAMES
487    </pre>
488    If this bit is set, names used to identify capturing subpatterns need not be
489    unique. This can be helpful for certain types of pattern when it is known that
490    only one instance of the named subpattern can ever be matched. There are more
491    details of named subpatterns below; see also the
492    <a href="pcrepattern.html"><b>pcrepattern</b></a>
493    documentation.
494  <pre>  <pre>
495    PCRE_EXTENDED    PCRE_EXTENDED
496  </PRE>  </pre>
 </P>  
 <P>  
497  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
498  ignored except when escaped or inside a character class. Whitespace does not  ignored except when escaped or inside a character class. Whitespace does not
499  include the VT character (code 11). In addition, characters between an  include the VT character (code 11). In addition, characters between an
500  unescaped # outside a character class and the next newline character,  unescaped # outside a character class and the next newline, inclusive, are also
501  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
502  be changed within a pattern by a (?x) option setting.  pattern by a (?x) option setting.
503  </P>  </P>
504  <P>  <P>
505  This option makes it possible to include comments inside complicated patterns.  This option makes it possible to include comments inside complicated patterns.
506  Note, however, that this applies only to data characters. Whitespace characters  Note, however, that this applies only to data characters. Whitespace characters
507  may never appear within special character sequences in a pattern, for example  may never appear within special character sequences in a pattern, for example
508  within the sequence (?( which introduces a conditional subpattern.  within the sequence (?( which introduces a conditional subpattern.
 </P>  
 <P>  
509  <pre>  <pre>
510    PCRE_EXTRA    PCRE_EXTRA
511  </PRE>  </pre>
 </P>  
 <P>  
512  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
513  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
514  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
515  special meaning causes an error, thus reserving these combinations for future  special meaning causes an error, thus reserving these combinations for future
516  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
517  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
518  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
519  pattern.  this option. It can also be set by a (?X) option setting within a pattern.
520  </P>  <pre>
521  <P>    PCRE_FIRSTLINE
522    </pre>
523    If this option is set, an unanchored pattern is required to match before or at
524    the first newline in the subject string, though the matched text may continue
525    over the newline.
526  <pre>  <pre>
527    PCRE_MULTILINE    PCRE_MULTILINE
528  </PRE>  </pre>
529  </P>  By default, PCRE treats the subject string as consisting of a single line of
530  <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"  
531  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
532  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
533  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 535  Perl.
535  </P>  </P>
536  <P>  <P>
537  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
538  match immediately following or immediately before any newline in the subject  match immediately following or immediately before internal newlines in the
539  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
540  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
541  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
542  occurrences of ^ or $ in a pattern, setting PCRE_MULTILINE has no effect.  occurrences of ^ or $ in a pattern, setting PCRE_MULTILINE has no effect.
543    <pre>
544      PCRE_NEWLINE_CR
545      PCRE_NEWLINE_LF
546      PCRE_NEWLINE_CRLF
547      PCRE_NEWLINE_ANY
548    </pre>
549    These options override the default newline definition that was chosen when PCRE
550    was built. Setting the first or the second specifies that a newline is
551    indicated by a single character (CR or LF, respectively). Setting
552    PCRE_NEWLINE_CRLF specifies that a newline is indicated by the two-character
553    CRLF sequence. Setting PCRE_NEWLINE_ANY specifies that any Unicode newline
554    sequence should be recognized. The Unicode newline sequences are the three just
555    mentioned, plus the single characters VT (vertical tab, U+000B), FF (formfeed,
556    U+000C), NEL (next line, U+0085), LS (line separator, U+2028), and PS
557    (paragraph separator, U+2029). The last two are recognized only in UTF-8 mode.
558    </P>
559    <P>
560    The newline setting in the options word uses three bits that are treated
561    as a number, giving eight possibilities. Currently only five are used (default
562    plus the four values above). This means that if you set more than one newline
563    option, the combination may or may not be sensible. For example,
564    PCRE_NEWLINE_CR with PCRE_NEWLINE_LF is equivalent to PCRE_NEWLINE_CRLF, but
565    other combinations yield unused numbers and cause an error.
566    </P>
567    <P>
568    The only time that a line break is specially recognized when compiling a
569    pattern is if PCRE_EXTENDED is set, and an unescaped # outside a character
570    class is encountered. This indicates a comment that lasts until after the next
571    line break sequence. In other circumstances, line break sequences are treated
572    as literal data, except that in PCRE_EXTENDED mode, both CR and LF are treated
573    as whitespace characters and are therefore ignored.
574  </P>  </P>
575  <P>  <P>
576    The newline option that is set at compile time becomes the default that is used
577    for <b>pcre_exec()</b> and <b>pcre_dfa_exec()</b>, but it can be overridden.
578  <pre>  <pre>
579    PCRE_NO_AUTO_CAPTURE    PCRE_NO_AUTO_CAPTURE
580  </PRE>  </pre>
 </P>  
 <P>  
581  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
582  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
583  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
584  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
585  in Perl.  in Perl.
 </P>  
 <P>  
586  <pre>  <pre>
587    PCRE_UNGREEDY    PCRE_UNGREEDY
588  </PRE>  </pre>
 </P>  
 <P>  
589  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
590  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
591  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>  
592  <pre>  <pre>
593    PCRE_UTF8    PCRE_UTF8
594  </PRE>  </pre>
 </P>  
 <P>  
595  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
596  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
597  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
598  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
599  behaviour of PCRE are given in the  behaviour of PCRE are given in the
600  <a href="pcre.html#utf8support">section on UTF-8 support</a>  <a href="pcre.html#utf8support">section on UTF-8 support</a>
601  in the main  in the main
602  <a href="pcre.html"><b>pcre</b></a>  <a href="pcre.html"><b>pcre</b></a>
603  page.  page.
604    <pre>
605      PCRE_NO_UTF8_CHECK
606    </pre>
607    When PCRE_UTF8 is set, the validity of the pattern as a UTF-8 string is
608    automatically checked. If an invalid UTF-8 sequence of bytes is found,
609    <b>pcre_compile()</b> returns an error. If you already know that your pattern is
610    valid, and you want to skip this check for performance reasons, you can set the
611    PCRE_NO_UTF8_CHECK option. When it is set, the effect of passing an invalid
612    UTF-8 string as a pattern is undefined. It may cause your program to crash.
613    Note that this option can also be passed to <b>pcre_exec()</b> and
614    <b>pcre_dfa_exec()</b>, to suppress the UTF-8 validity checking of subject
615    strings.
616    </P>
617    <br><a name="SEC8" href="#TOC1">COMPILATION ERROR CODES</a><br>
618    <P>
619    The following table lists the error codes than may be returned by
620    <b>pcre_compile2()</b>, along with the error messages that may be returned by
621    both compiling functions. As PCRE has developed, some error codes have fallen
622    out of use. To avoid confusion, they have not been re-used.
623    <pre>
624       0  no error
625       1  \ at end of pattern
626       2  \c at end of pattern
627       3  unrecognized character follows \
628       4  numbers out of order in {} quantifier
629       5  number too big in {} quantifier
630       6  missing terminating ] for character class
631       7  invalid escape sequence in character class
632       8  range out of order in character class
633       9  nothing to repeat
634      10  [this code is not in use]
635      11  internal error: unexpected repeat
636      12  unrecognized character after (?
637      13  POSIX named classes are supported only within a class
638      14  missing )
639      15  reference to non-existent subpattern
640      16  erroffset passed as NULL
641      17  unknown option bit(s) set
642      18  missing ) after comment
643      19  [this code is not in use]
644      20  regular expression too large
645      21  failed to get memory
646      22  unmatched parentheses
647      23  internal error: code overflow
648      24  unrecognized character after (?&#60;
649      25  lookbehind assertion is not fixed length
650      26  malformed number or name after (?(
651      27  conditional group contains more than two branches
652      28  assertion expected after (?(
653      29  (?R or (?digits must be followed by )
654      30  unknown POSIX class name
655      31  POSIX collating elements are not supported
656      32  this version of PCRE is not compiled with PCRE_UTF8 support
657      33  [this code is not in use]
658      34  character value in \x{...} sequence is too large
659      35  invalid condition (?(0)
660      36  \C not allowed in lookbehind assertion
661      37  PCRE does not support \L, \l, \N, \U, or \u
662      38  number after (?C is &#62; 255
663      39  closing ) for (?C expected
664      40  recursive call could loop indefinitely
665      41  unrecognized character after (?P
666      42  syntax error in subpattern name (missing terminator)
667      43  two named subpatterns have the same name
668      44  invalid UTF-8 string
669      45  support for \P, \p, and \X has not been compiled
670      46  malformed \P or \p sequence
671      47  unknown property name after \P or \p
672      48  subpattern name is too long (maximum 32 characters)
673      49  too many named subpatterns (maximum 10,000)
674      50  repeated subpattern is too long
675      51  octal value is greater than \377 (not in UTF-8 mode)
676      52  internal error: overran compiling workspace
677      53  internal error: previously-checked referenced subpattern not found
678      54  DEFINE group contains more than one branch
679      55  repeating a DEFINE group is not allowed
680      56  inconsistent NEWLINE options"
681    </PRE>
682  </P>  </P>
683  <br><a name="SEC6" href="#TOC1">STUDYING A PATTERN</a><br>  <br><a name="SEC9" href="#TOC1">STUDYING A PATTERN</a><br>
684  <P>  <P>
685  <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>
686  <b>const char **<i>errptr</i>);</b>  <b>const char **<i>errptr</i>);</b>
687  </P>  </P>
688  <P>  <P>
689  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
690  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
691  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
692  argument. If studing the pattern produces additional information that will help  argument. If studying the pattern produces additional information that will
693  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
694  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
695    results of the study.
696  </P>  </P>
697  <P>  <P>
698  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
699  <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
700  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
701  described below. If studying the pattern does not produce any additional  described
702  information, <b>pcre_study()</b> returns NULL. In that circumstance, if the  <a href="#extradata">below</a>
703  calling program wants to pass some of the other fields to <b>pcre_exec()</b>, it  in the section on matching a pattern.
704  must set up its own <b>pcre_extra</b> block.  </P>
705    <P>
706    If studying the pattern does not produce any additional information
707    <b>pcre_study()</b> returns NULL. In that circumstance, if the calling program
708    wants to pass any of the other fields to <b>pcre_exec()</b>, it must set up its
709    own <b>pcre_extra</b> block.
710  </P>  </P>
711  <P>  <P>
712  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
713  for <b>pcre_study()</b>, and this argument should always be zero.  options are defined, and this argument should always be zero.
714  </P>  </P>
715  <P>  <P>
716  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
717  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
718  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
719  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
720  be sure that it has run successfully.  should test the error pointer for NULL after calling <b>pcre_study()</b>, to be
721    sure that it has run successfully.
722  </P>  </P>
723  <P>  <P>
724  This is a typical call to <b>pcre_study</b>():  This is a typical call to <b>pcre_study</b>():
 </P>  
 <P>  
725  <pre>  <pre>
726    pcre_extra *pe;    pcre_extra *pe;
727    pe = pcre_study(    pe = pcre_study(
728      re,             /* result of pcre_compile() */      re,             /* result of pcre_compile() */
729      0,              /* no options exist */      0,              /* no options exist */
730      &error);        /* set to NULL or points to a message */      &error);        /* set to NULL or points to a message */
731  </PRE>  </pre>
 </P>  
 <P>  
732  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
733  not have a single fixed starting character. A bitmap of possible starting  not have a single fixed starting character. A bitmap of possible starting
734  characters is created.  bytes is created.
735  </P>  <a name="localesupport"></a></P>
736  <a name="localesupport"></a><br><a name="SEC7" href="#TOC1">LOCALE SUPPORT</a><br>  <br><a name="SEC10" href="#TOC1">LOCALE SUPPORT</a><br>
737  <P>  <P>
738  PCRE handles caseless matching, and determines whether characters are letters,  PCRE handles caseless matching, and determines whether characters are letters
739  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
740  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
741  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
742  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
743  is NULL, and is sufficient for many applications.  support. The use of locales with Unicode is discouraged.
744  </P>  </P>
745  <P>  <P>
746  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
747  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,
748  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,
749  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
750  French locale (where accented characters with codes greater than 128 are  default. As more and more applications change to using Unicode, the need for
751  treated as letters), the following code could be used:  this locale support is expected to die away.
752  </P>  </P>
753  <P>  <P>
754    External tables are built by calling the <b>pcre_maketables()</b> function,
755    which has no arguments, in the relevant locale. The result can then be passed
756    to <b>pcre_compile()</b> or <b>pcre_exec()</b> as often as necessary. For
757    example, to build and use tables that are appropriate for the French locale
758    (where accented characters with values greater than 128 are treated as letters),
759    the following code could be used:
760  <pre>  <pre>
761    setlocale(LC_CTYPE, "fr");    setlocale(LC_CTYPE, "fr_FR");
762    tables = pcre_maketables();    tables = pcre_maketables();
763    re = pcre_compile(..., tables);    re = pcre_compile(..., tables);
764  </PRE>  </pre>
765    When <b>pcre_maketables()</b> runs, the tables are built in memory that is
766    obtained via <b>pcre_malloc</b>. It is the caller's responsibility to ensure
767    that the memory containing the tables remains available for as long as it is
768    needed.
769  </P>  </P>
770  <P>  <P>
771  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  
772  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>
773  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
774  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
775  in different locales. It is the caller's responsibility to ensure that the  different patterns can be compiled in different locales.
 memory containing the tables remains available for as long as it is needed.  
776  </P>  </P>
777  <br><a name="SEC8" href="#TOC1">INFORMATION ABOUT A PATTERN</a><br>  <P>
778    It is possible to pass a table pointer or NULL (indicating the use of the
779    internal tables) to <b>pcre_exec()</b>. Although not intended for this purpose,
780    this facility could be used to match a pattern in a different locale from the
781    one in which it was compiled. Passing table pointers at run time is discussed
782    below in the section on matching a pattern.
783    </P>
784    <br><a name="SEC11" href="#TOC1">INFORMATION ABOUT A PATTERN</a><br>
785  <P>  <P>
786  <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>
787  <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 798  the pattern was not studied. The third a
798  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
799  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
800  the following negative numbers:  the following negative numbers:
 </P>  
 <P>  
801  <pre>  <pre>
802    PCRE_ERROR_NULL       the argument <i>code</i> was NULL    PCRE_ERROR_NULL       the argument <i>code</i> was NULL
803                          the argument <i>where</i> was NULL                          the argument <i>where</i> was NULL
804    PCRE_ERROR_BADMAGIC   the "magic number" was not found    PCRE_ERROR_BADMAGIC   the "magic number" was not found
805    PCRE_ERROR_BADOPTION  the value of <i>what</i> was invalid    PCRE_ERROR_BADOPTION  the value of <i>what</i> was invalid
806  </PRE>  </pre>
807  </P>  The "magic number" is placed at the start of each compiled pattern as an simple
808  <P>  check against passing an arbitrary memory pointer. Here is a typical call of
809  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>  
810  <pre>  <pre>
811    int rc;    int rc;
812    unsigned long int length;    size_t length;
813    rc = pcre_fullinfo(    rc = pcre_fullinfo(
814      re,               /* result of pcre_compile() */      re,               /* result of pcre_compile() */
815      pe,               /* result of pcre_study(), or NULL */      pe,               /* result of pcre_study(), or NULL */
816      PCRE_INFO_SIZE,   /* what is required */      PCRE_INFO_SIZE,   /* what is required */
817      &length);         /* where to put the data */      &length);         /* where to put the data */
818  </PRE>  </pre>
 </P>  
 <P>  
819  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
820  as follows:  as follows:
 </P>  
 <P>  
821  <pre>  <pre>
822    PCRE_INFO_BACKREFMAX    PCRE_INFO_BACKREFMAX
823  </PRE>  </pre>
 </P>  
 <P>  
824  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
825  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
826  no back references.  no back references.
 </P>  
 <P>  
827  <pre>  <pre>
828    PCRE_INFO_CAPTURECOUNT    PCRE_INFO_CAPTURECOUNT
829  </PRE>  </pre>
 </P>  
 <P>  
830  Return the number of capturing subpatterns in the pattern. The fourth argument  Return the number of capturing subpatterns in the pattern. The fourth argument
831  should point to an \fbint\fR variable.  should point to an <b>int</b> variable.
832  </P>  <pre>
833  <P>    PCRE_INFO_DEFAULT_TABLES
834    </pre>
835    Return a pointer to the internal default character tables within PCRE. The
836    fourth argument should point to an <b>unsigned char *</b> variable. This
837    information call is provided for internal use by the <b>pcre_study()</b>
838    function. External callers can cause PCRE to use its internal tables by passing
839    a NULL table pointer.
840  <pre>  <pre>
841    PCRE_INFO_FIRSTBYTE    PCRE_INFO_FIRSTBYTE
842  </PRE>  </pre>
 </P>  
 <P>  
843  Return information about the first byte of any matched string, for a  Return information about the first byte of any matched string, for a
844  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>
845  old name is still recognized for backwards compatibility.)  variable. (This option used to be called PCRE_INFO_FIRSTCHAR; the old name is
846  </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  
847  </P>  </P>
848  <P>  <P>
849    If there is a fixed first byte, for example, from a pattern such as
850    (cat|cow|coyote), its value is returned. Otherwise, if either
851    <br>
852    <br>
853  (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
854  starts with "^", or  starts with "^", or
855  </P>  <br>
856  <P>  <br>
857  (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
858  (if it were set, the pattern would be anchored),  (if it were set, the pattern would be anchored),
859  </P>  <br>
860  <P>  <br>
861  -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
862  subject string or after any newline within the string. Otherwise -2 is  subject string or after any newline within the string. Otherwise -2 is
863  returned. For anchored patterns, -2 is returned.  returned. For anchored patterns, -2 is returned.
 </P>  
 <P>  
864  <pre>  <pre>
865    PCRE_INFO_FIRSTTABLE    PCRE_INFO_FIRSTTABLE
866  </PRE>  </pre>
 </P>  
 <P>  
867  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
868  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
869  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
870  fourth argument should point to an <b>unsigned char *</b> variable.  fourth argument should point to an <b>unsigned char *</b> variable.
 </P>  
 <P>  
871  <pre>  <pre>
872    PCRE_INFO_LASTLITERAL    PCRE_INFO_LASTLITERAL
873  </PRE>  </pre>
 </P>  
 <P>  
874  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
875  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
876  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 878  returned. For anchored patterns, a last
878  follows something of variable length. For example, for the pattern  follows something of variable length. For example, for the pattern
879  /^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
880  is -1.  is -1.
 </P>  
 <P>  
881  <pre>  <pre>
882    PCRE_INFO_NAMECOUNT    PCRE_INFO_NAMECOUNT
883    PCRE_INFO_NAMEENTRYSIZE    PCRE_INFO_NAMEENTRYSIZE
884    PCRE_INFO_NAMETABLE    PCRE_INFO_NAMETABLE
885  </PRE>  </pre>
 </P>  
 <P>  
886  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
887  names are just an additional way of identifying the parentheses, which still  names are just an additional way of identifying the parentheses, which still
888  acquire a number. A caller that wants to extract data from a named subpattern  acquire numbers. Several convenience functions such as
889  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
890  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
891  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
892  table for the pattern.  output vector (described with <b>pcre_exec()</b> below). To do the conversion,
893    you need to use the name-to-number map, which is described by these three
894    values.
895  </P>  </P>
896  <P>  <P>
897  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 901  length of the longest name. PCRE_INFO_NA
901  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
902  are the number of the capturing parenthesis, most significant byte first. The  are the number of the capturing parenthesis, most significant byte first. The
903  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
904  alphabetical order. For example, consider the following pattern (assume  alphabetical order. When PCRE_DUPNAMES is set, duplicate names are in order of
905    their parentheses numbers. For example, consider the following pattern (assume
906  PCRE_EXTENDED is set, so white space - including newlines - is ignored):  PCRE_EXTENDED is set, so white space - including newlines - is ignored):
 </P>  
 <P>  
907  <pre>  <pre>
908    (?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) )
909    (?P&#60;month&#62;\d\d) - (?P&#60;day&#62;\d\d) )  </pre>
 </PRE>  
 </P>  
 <P>  
910  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
911  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
912  bytes shows in hex, and undefined bytes shown as ??:  bytes shows in hexadecimal, and undefined bytes shown as ??:
 </P>  
 <P>  
913  <pre>  <pre>
914    00 01 d  a  t  e  00 ??    00 01 d  a  t  e  00 ??
915    00 05 d  a  y  00 ?? ??    00 05 d  a  y  00 ?? ??
916    00 04 m  o  n  t  h  00    00 04 m  o  n  t  h  00
917    00 02 y  e  a  r  00 ??    00 02 y  e  a  r  00 ??
918  </PRE>  </pre>
919  </P>  When writing code to extract data from named subpatterns using the
920  <P>  name-to-number map, remember that the length of the entries is likely to be
921  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>  
922  <pre>  <pre>
923    PCRE_INFO_OPTIONS    PCRE_INFO_OPTIONS
924  </PRE>  </pre>
 </P>  
 <P>  
925  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
926  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
927  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 930  top-level option settings within the pat
930  <P>  <P>
931  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
932  alternatives begin with one of the following:  alternatives begin with one of the following:
 </P>  
 <P>  
933  <pre>  <pre>
934    ^     unless PCRE_MULTILINE is set    ^     unless PCRE_MULTILINE is set
935    \A    always    \A    always
936    \G    always    \G    always
937    .*    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
938            references to the subpattern in which .* appears  </pre>
 </PRE>  
 </P>  
 <P>  
939  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
940  <b>pcre_fullinfo()</b>.  <b>pcre_fullinfo()</b>.
 </P>  
 <P>  
941  <pre>  <pre>
942    PCRE_INFO_SIZE    PCRE_INFO_SIZE
943  </PRE>  </pre>
 </P>  
 <P>  
944  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
945  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
946  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>
947  variable.  variable.
 </P>  
 <P>  
948  <pre>  <pre>
949    PCRE_INFO_STUDYSIZE    PCRE_INFO_STUDYSIZE
950  </PRE>  </pre>
951  </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  
952  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
953  <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
954  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
955  <b>size_t</b> variable.  <b>size_t</b> variable.
956  </P>  </P>
957  <br><a name="SEC9" href="#TOC1">OBSOLETE INFO FUNCTION</a><br>  <br><a name="SEC12" href="#TOC1">OBSOLETE INFO FUNCTION</a><br>
958  <P>  <P>
959  <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>
960  <b>*<i>firstcharptr</i>);</b>  <b>*<i>firstcharptr</i>);</b>
# Line 753  restrictive to return all the available Line 965  restrictive to return all the available
965  programs should use <b>pcre_fullinfo()</b> instead. The yield of  programs should use <b>pcre_fullinfo()</b> instead. The yield of
966  <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
967  following negative numbers:  following negative numbers:
 </P>  
 <P>  
968  <pre>  <pre>
969    PCRE_ERROR_NULL       the argument <i>code</i> was NULL    PCRE_ERROR_NULL       the argument <i>code</i> was NULL
970    PCRE_ERROR_BADMAGIC   the "magic number" was not found    PCRE_ERROR_BADMAGIC   the "magic number" was not found
971  </PRE>  </pre>
 </P>  
 <P>  
972  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
973  pattern was compiled is placed in the integer it points to (see  pattern was compiled is placed in the integer it points to (see
974  PCRE_INFO_OPTIONS above).  PCRE_INFO_OPTIONS above).
# Line 770  If the pattern is not anchored and the < Line 978  If the pattern is not anchored and the <
978  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
979  string (see PCRE_INFO_FIRSTBYTE above).  string (see PCRE_INFO_FIRSTBYTE above).
980  </P>  </P>
981  <br><a name="SEC10" href="#TOC1">MATCHING A PATTERN</a><br>  <br><a name="SEC13" href="#TOC1">REFERENCE COUNTS</a><br>
982    <P>
983    <b>int pcre_refcount(pcre *<i>code</i>, int <i>adjust</i>);</b>
984    </P>
985    <P>
986    The <b>pcre_refcount()</b> function is used to maintain a reference count in the
987    data block that contains a compiled pattern. It is provided for the benefit of
988    applications that operate in an object-oriented manner, where different parts
989    of the application may be using the same compiled pattern, but you want to free
990    the block when they are all done.
991    </P>
992    <P>
993    When a pattern is compiled, the reference count field is initialized to zero.
994    It is changed only by calling this function, whose action is to add the
995    <i>adjust</i> value (which may be positive or negative) to it. The yield of the
996    function is the new value. However, the value of the count is constrained to
997    lie between 0 and 65535, inclusive. If the new value is outside these limits,
998    it is forced to the appropriate limit value.
999    </P>
1000    <P>
1001    Except when it is zero, the reference count is not correctly preserved if a
1002    pattern is compiled on one host and then transferred to a host whose byte-order
1003    is different. (This seems a highly unlikely scenario.)
1004    </P>
1005    <br><a name="SEC14" href="#TOC1">MATCHING A PATTERN: THE TRADITIONAL FUNCTION</a><br>
1006  <P>  <P>
1007  <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>
1008  <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 1010  string (see PCRE_INFO_FIRSTBYTE above).
1010  </P>  </P>
1011  <P>  <P>
1012  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
1013  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
1014  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
1015  <i>extra</i> argument.  <i>extra</i> argument. This function is the main matching facility of the
1016    library, and it operates in a Perl-like manner. For specialist use there is
1017    also an alternative matching function, which is described
1018    <a href="#dfamatch">below</a>
1019    in the section about the <b>pcre_dfa_exec()</b> function.
1020  </P>  </P>
1021  <P>  <P>
1022  Here is an example of a simple call to <b>pcre_exec()</b>:  In most applications, the pattern will have been compiled (and optionally
1023    studied) in the same process that calls <b>pcre_exec()</b>. However, it is
1024    possible to save compiled patterns and study data, and then use them later
1025    in different processes, possibly even on different hosts. For a discussion
1026    about this, see the
1027    <a href="pcreprecompile.html"><b>pcreprecompile</b></a>
1028    documentation.
1029  </P>  </P>
1030  <P>  <P>
1031    Here is an example of a simple call to <b>pcre_exec()</b>:
1032  <pre>  <pre>
1033    int rc;    int rc;
1034    int ovector[30];    int ovector[30];
# Line 796  Here is an example of a simple call to < Line 1039  Here is an example of a simple call to <
1039      11,             /* the length of the subject string */      11,             /* the length of the subject string */
1040      0,              /* start at offset 0 in the subject */      0,              /* start at offset 0 in the subject */
1041      0,              /* default options */      0,              /* default options */
1042      ovector,        /* vector for substring information */      ovector,        /* vector of integers for substring information */
1043      30);            /* number of elements in the vector */      30);            /* number of elements (NOT size in bytes) */
1044  </PRE>  <a name="extradata"></a></PRE>
1045  </P>  </P>
1046    <br><b>
1047    Extra data for <b>pcre_exec()</b>
1048    </b><br>
1049  <P>  <P>
1050  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>
1051  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
1052  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
1053  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
1054  </P>  fields (not necessarily in this order):
 <P>  
1055  <pre>  <pre>
1056    unsigned long int <i>flags</i>;    unsigned long int <i>flags</i>;
1057    void *<i>study_data</i>;    void *<i>study_data</i>;
1058    unsigned long int <i>match_limit</i>;    unsigned long int <i>match_limit</i>;
1059      unsigned long int <i>match_limit_recursion</i>;
1060    void *<i>callout_data</i>;    void *<i>callout_data</i>;
1061  </PRE>    const unsigned char *<i>tables</i>;
1062  </P>  </pre>
 <P>  
1063  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
1064  are set. The flag bits are:  are set. The flag bits are:
 </P>  
 <P>  
1065  <pre>  <pre>
1066    PCRE_EXTRA_STUDY_DATA    PCRE_EXTRA_STUDY_DATA
1067    PCRE_EXTRA_MATCH_LIMIT    PCRE_EXTRA_MATCH_LIMIT
1068      PCRE_EXTRA_MATCH_LIMIT_RECURSION
1069    PCRE_EXTRA_CALLOUT_DATA    PCRE_EXTRA_CALLOUT_DATA
1070  </PRE>    PCRE_EXTRA_TABLES
1071  </P>  </pre>
 <P>  
1072  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
1073  <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
1074  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
1075  the block by setting the other fields.  the block by setting the other fields and their corresponding flag bits.
1076  </P>  </P>
1077  <P>  <P>
1078  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
1079  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,
1080  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
1081  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.  
1082  </P>  </P>
1083  <P>  <P>
1084  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
1085  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
1086  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
1087  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
1088  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
1089  exceeded, <b>pcre_exec()</b> returns PCRE_ERROR_MATCHLIMIT.  string.
1090  </P>  </P>
1091  <P>  <P>
1092  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
1093  which is described in the <b>pcrecallout</b> documentation.  default is 10 million, which handles all but the most extreme cases. You can
1094  </P>  override the default by suppling <b>pcre_exec()</b> with a <b>pcre_extra</b>
1095  <P>  block in which <i>match_limit</i> is set, and PCRE_EXTRA_MATCH_LIMIT is set in
1096  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
1097  unused bits must be zero. This limits <b>pcre_exec()</b> to matching at the  PCRE_ERROR_MATCHLIMIT.
1098  first matching position. However, if a pattern was compiled with PCRE_ANCHORED,  </P>
1099  or turned out to be anchored by virtue of its contents, it cannot be made  <P>
1100  unachored at matching time.  The <i>match_limit_recursion</i> field is similar to <i>match_limit</i>, but
1101    instead of limiting the total number of times that <b>match()</b> is called, it
1102    limits the depth of recursion. The recursion depth is a smaller number than the
1103    total number of calls, because not all calls to <b>match()</b> are recursive.
1104    This limit is of use only if it is set smaller than <i>match_limit</i>.
1105    </P>
1106    <P>
1107    Limiting the recursion depth limits the amount of stack that can be used, or,
1108    when PCRE has been compiled to use memory on the heap instead of the stack, the
1109    amount of heap memory that can be used.
1110    </P>
1111    <P>
1112    The default value for <i>match_limit_recursion</i> can be set when PCRE is
1113    built; the default default is the same value as the default for
1114    <i>match_limit</i>. You can override the default by suppling <b>pcre_exec()</b>
1115    with a <b>pcre_extra</b> block in which <i>match_limit_recursion</i> is set, and
1116    PCRE_EXTRA_MATCH_LIMIT_RECURSION is set in the <i>flags</i> field. If the limit
1117    is exceeded, <b>pcre_exec()</b> returns PCRE_ERROR_RECURSIONLIMIT.
1118  </P>  </P>
1119  <P>  <P>
1120  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,
1121    which is described in the
1122    <a href="pcrecallout.html"><b>pcrecallout</b></a>
1123    documentation.
1124  </P>  </P>
1125  <P>  <P>
1126    The <i>tables</i> field is used to pass a character tables pointer to
1127    <b>pcre_exec()</b>; this overrides the value that is stored with the compiled
1128    pattern. A non-NULL value is stored with the compiled pattern only if custom
1129    tables were supplied to <b>pcre_compile()</b> via its <i>tableptr</i> argument.
1130    If NULL is passed to <b>pcre_exec()</b> using this mechanism, it forces PCRE's
1131    internal tables to be used. This facility is helpful when re-using patterns
1132    that have been saved after compiling with an external set of tables, because
1133    the external tables might be at a different address when <b>pcre_exec()</b> is
1134    called. See the
1135    <a href="pcreprecompile.html"><b>pcreprecompile</b></a>
1136    documentation for a discussion of saving compiled patterns for later use.
1137    </P>
1138    <br><b>
1139    Option bits for <b>pcre_exec()</b>
1140    </b><br>
1141    <P>
1142    The unused bits of the <i>options</i> argument for <b>pcre_exec()</b> must be
1143    zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_<i>xxx</i>,
1144    PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NO_UTF8_CHECK and PCRE_PARTIAL.
1145    <pre>
1146      PCRE_ANCHORED
1147    </pre>
1148    The PCRE_ANCHORED option limits <b>pcre_exec()</b> to matching at the first
1149    matching position. If a pattern was compiled with PCRE_ANCHORED, or turned out
1150    to be anchored by virtue of its contents, it cannot be made unachored at
1151    matching time.
1152    <pre>
1153      PCRE_NEWLINE_CR
1154      PCRE_NEWLINE_LF
1155      PCRE_NEWLINE_CRLF
1156      PCRE_NEWLINE_ANY
1157    </pre>
1158    These options override the newline definition that was chosen or defaulted when
1159    the pattern was compiled. For details, see the description of
1160    <b>pcre_compile()</b> above. During matching, the newline choice affects the
1161    behaviour of the dot, circumflex, and dollar metacharacters. It may also alter
1162    the way the match position is advanced after a match failure for an unanchored
1163    pattern. When PCRE_NEWLINE_CRLF or PCRE_NEWLINE_ANY is set, and a match attempt
1164    fails when the current position is at a CRLF sequence, the match position is
1165    advanced by two characters instead of one, in other words, to after the CRLF.
1166  <pre>  <pre>
1167    PCRE_NOTBOL    PCRE_NOTBOL
1168  </PRE>  </pre>
1169  </P>  This option specifies that first character of the subject string is not the
1170  <P>  beginning of a line, so the circumflex metacharacter should not match before
1171  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
1172  circumflex metacharacter should not match before it. Setting this without  never to match. This option affects only the behaviour of the circumflex
1173  PCRE_MULTILINE (at compile time) causes circumflex never to match.  metacharacter. It does not affect \A.
 </P>  
 <P>  
1174  <pre>  <pre>
1175    PCRE_NOTEOL    PCRE_NOTEOL
1176  </PRE>  </pre>
1177  </P>  This option specifies that the end of the subject string is not the end of a
1178  <P>  line, so the dollar metacharacter should not match it nor (except in multiline
1179  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
1180  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
1181  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>  
1182  <pre>  <pre>
1183    PCRE_NOTEMPTY    PCRE_NOTEMPTY
1184  </PRE>  </pre>
 </P>  
 <P>  
1185  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
1186  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
1187  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>  
1188  <pre>  <pre>
1189    a?b?    a?b?
1190  </PRE>  </pre>
 </P>  
 <P>  
1191  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
1192  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
1193  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 1197  Perl has no direct equivalent of PCRE_NO
1197  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
1198  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
1199  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
1200  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
1201  below) and trying an ordinary match again.  starting offset (see below) and trying an ordinary match again. There is some
1202    code that demonstrates how to do this in the <i>pcredemo.c</i> sample program.
1203    <pre>
1204      PCRE_NO_UTF8_CHECK
1205    </pre>
1206    When PCRE_UTF8 is set at compile time, the validity of the subject as a UTF-8
1207    string is automatically checked when <b>pcre_exec()</b> is subsequently called.
1208    The value of <i>startoffset</i> is also checked to ensure that it points to the
1209    start of a UTF-8 character. If an invalid UTF-8 sequence of bytes is found,
1210    <b>pcre_exec()</b> returns the error PCRE_ERROR_BADUTF8. If <i>startoffset</i>
1211    contains an invalid value, PCRE_ERROR_BADUTF8_OFFSET is returned.
1212    </P>
1213    <P>
1214    If you already know that your subject is valid, and you want to skip these
1215    checks for performance reasons, you can set the PCRE_NO_UTF8_CHECK option when
1216    calling <b>pcre_exec()</b>. You might want to do this for the second and
1217    subsequent calls to <b>pcre_exec()</b> if you are making repeated calls to find
1218    all the matches in a single subject string. However, you should be sure that
1219    the value of <i>startoffset</i> points to the start of a UTF-8 character. When
1220    PCRE_NO_UTF8_CHECK is set, the effect of passing an invalid UTF-8 string as a
1221    subject, or a value of <i>startoffset</i> that does not point to the start of a
1222    UTF-8 character, is undefined. Your program may crash.
1223    <pre>
1224      PCRE_PARTIAL
1225    </pre>
1226    This option turns on the partial matching feature. If the subject string fails
1227    to match the pattern, but at some point during the matching process the end of
1228    the subject was reached (that is, the subject partially matches the pattern and
1229    the failure to match occurred only because there were not enough subject
1230    characters), <b>pcre_exec()</b> returns PCRE_ERROR_PARTIAL instead of
1231    PCRE_ERROR_NOMATCH. When PCRE_PARTIAL is used, there are restrictions on what
1232    may appear in the pattern. These are discussed in the
1233    <a href="pcrepartial.html"><b>pcrepartial</b></a>
1234    documentation.
1235  </P>  </P>
1236    <br><b>
1237    The string to be matched by <b>pcre_exec()</b>
1238    </b><br>
1239  <P>  <P>
1240  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
1241  <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
1242  <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
1243  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
1244  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
1245  </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.  
1246  </P>  </P>
1247  <P>  <P>
1248  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 1250  same subject by calling <b>pcre_exec()</
1250  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
1251  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
1252  lookbehind. For example, consider the pattern  lookbehind. For example, consider the pattern
 </P>  
 <P>  
1253  <pre>  <pre>
1254    \Biss\B    \Biss\B
1255  </PRE>  </pre>
 </P>  
 <P>  
1256  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
1257  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
1258  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 1265  behind the starting point to discover th
1265  </P>  </P>
1266  <P>  <P>
1267  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
1268  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
1269  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.
1270  </P>  </P>
1271    <br><b>
1272    How <b>pcre_exec()</b> returns captured substrings
1273    </b><br>
1274  <P>  <P>
1275  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
1276  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 1282  kinds of parenthesized subpattern that d
1282  <P>  <P>
1283  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
1284  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
1285  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>:
1286  back captured substrings, each substring using a pair of integers. The  this argument is NOT the size of <i>ovector</i> in bytes.
1287  remaining third of the vector is used as workspace by <b>pcre_exec()</b> while  </P>
1288  matching capturing subpatterns, and is not available for passing back  <P>
1289  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,
1290  three. If it is not, it is rounded down.  each substring using a pair of integers. The remaining third of the vector is
1291    used as workspace by <b>pcre_exec()</b> while matching capturing subpatterns,
1292    and is not available for passing back information. The length passed in
1293    <i>ovecsize</i> should always be a multiple of three. If it is not, it is
1294    rounded down.
1295  </P>  </P>
1296  <P>  <P>
1297  When a match has been successful, information about captured substrings is  When a match is successful, information about captured substrings is returned
1298  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
1299  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
1300  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
1301  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
1302  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
1303  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
1304  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>
1305  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
1306  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
1307  just the first pair of offsets has been set.  capturing subpatterns, the return value from a successful match is 1,
1308  </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.  
1309  </P>  </P>
1310  <P>  <P>
1311  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
1312  string that it matched that gets returned.  string that it matched that is returned.
1313  </P>  </P>
1314  <P>  <P>
1315  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
1316  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
1317  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
1318  <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
1319  <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
1320  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
1321  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
1322  to supply an <i>ovector</i>.  advisable to supply an <i>ovector</i>.
1323  </P>  </P>
1324  <P>  <P>
1325  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
1326  subpatterns there are in a compiled pattern. The smallest size for  subpatterns there are in a compiled pattern. The smallest size for
1327  <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
1328  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.
1329  </P>  </P>
1330  <P>  <P>
1331  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
1332  defined in the header file:  the subject when subpattern <i>n</i> has not been used at all. For example, if
1333    the string "abc" is matched against the pattern (a|(z))(bc) the return from the
1334    function is 4, and subpatterns 1 and 3 are matched, but 2 is not. When this
1335    happens, both values in the offset pairs corresponding to unused subpatterns
1336    are set to -1.
1337  </P>  </P>
1338  <P>  <P>
1339  <pre>  Offset values that correspond to unused subpatterns at the end of the
1340    PCRE_ERROR_NOMATCH        (-1)  expression are also set to -1. For example, if the string "abc" is matched
1341  </PRE>  against the pattern (abc)(x(yz)?)? subpatterns 2 and 3 are not matched. The
1342    return from the function is 2, because the highest used capturing subpattern
1343    number is 1. However, you can refer to the offsets for the second and third
1344    capturing subpatterns if you wish (assuming the vector is large enough, of
1345    course).
1346  </P>  </P>
1347  <P>  <P>
1348  The subject string did not match the pattern.  Some convenience functions are provided for extracting the captured substrings
1349  </P>  as separate strings. These are described below.
1350    <a name="errorlist"></a></P>
1351    <br><b>
1352    Error return values from <b>pcre_exec()</b>
1353    </b><br>
1354  <P>  <P>
1355    If <b>pcre_exec()</b> fails, it returns a negative number. The following are
1356    defined in the header file:
1357    <pre>
1358      PCRE_ERROR_NOMATCH        (-1)
1359    </pre>
1360    The subject string did not match the pattern.
1361  <pre>  <pre>
1362    PCRE_ERROR_NULL           (-2)    PCRE_ERROR_NULL           (-2)
1363  </PRE>  </pre>
 </P>  
 <P>  
1364  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
1365  NULL and <i>ovecsize</i> was not zero.  NULL and <i>ovecsize</i> was not zero.
 </P>  
 <P>  
1366  <pre>  <pre>
1367    PCRE_ERROR_BADOPTION      (-3)    PCRE_ERROR_BADOPTION      (-3)
1368  </PRE>  </pre>
 </P>  
 <P>  
1369  An unrecognized bit was set in the <i>options</i> argument.  An unrecognized bit was set in the <i>options</i> argument.
 </P>  
 <P>  
1370  <pre>  <pre>
1371    PCRE_ERROR_BADMAGIC       (-4)    PCRE_ERROR_BADMAGIC       (-4)
1372  </PRE>  </pre>
 </P>  
 <P>  
1373  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
1374  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
1375  magic number isn't present.  compiled in an environment of one endianness is run in an environment with the
1376  </P>  other endianness. This is the error that PCRE gives when the magic number is
1377  <P>  not present.
1378  <pre>  <pre>
1379    PCRE_ERROR_UNKNOWN_NODE   (-5)    PCRE_ERROR_UNKNOWN_OPCODE (-5)
1380  </PRE>  </pre>
 </P>  
 <P>  
1381  While running the pattern match, an unknown item was encountered in the  While running the pattern match, an unknown item was encountered in the
1382  compiled pattern. This error could be caused by a bug in PCRE or by overwriting  compiled pattern. This error could be caused by a bug in PCRE or by overwriting
1383  of the compiled pattern.  of the compiled pattern.
 </P>  
 <P>  
1384  <pre>  <pre>
1385    PCRE_ERROR_NOMEMORY       (-6)    PCRE_ERROR_NOMEMORY       (-6)
1386  </PRE>  </pre>
 </P>  
 <P>  
1387  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
1388  <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
1389  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
1390  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
1391  the end of matching.  automatically freed at the end of matching.
 </P>  
 <P>  
1392  <pre>  <pre>
1393    PCRE_ERROR_NOSUBSTRING    (-7)    PCRE_ERROR_NOSUBSTRING    (-7)
1394  </PRE>  </pre>
 </P>  
 <P>  
1395  This error is used by the <b>pcre_copy_substring()</b>,  This error is used by the <b>pcre_copy_substring()</b>,
1396  <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
1397  below). It is never returned by <b>pcre_exec()</b>.  below). It is never returned by <b>pcre_exec()</b>.
 </P>  
 <P>  
1398  <pre>  <pre>
1399    PCRE_ERROR_MATCHLIMIT     (-8)    PCRE_ERROR_MATCHLIMIT     (-8)
1400  </PRE>  </pre>
1401  </P>  The backtracking limit, as specified by the <i>match_limit</i> field in a
1402  <P>  <b>pcre_extra</b> structure (or defaulted) was reached. See the description
1403  The recursion and backtracking limit, as specified by the <i>match_limit</i>  above.
1404    <pre>
1405      PCRE_ERROR_CALLOUT        (-9)
1406    </pre>
1407    This error is never generated by <b>pcre_exec()</b> itself. It is provided for
1408    use by callout functions that want to yield a distinctive error code. See the
1409    <a href="pcrecallout.html"><b>pcrecallout</b></a>
1410    documentation for details.
1411    <pre>
1412      PCRE_ERROR_BADUTF8        (-10)
1413    </pre>
1414    A string that contains an invalid UTF-8 byte sequence was passed as a subject.
1415    <pre>
1416      PCRE_ERROR_BADUTF8_OFFSET (-11)
1417    </pre>
1418    The UTF-8 byte sequence that was passed as a subject was valid, but the value
1419    of <i>startoffset</i> did not point to the beginning of a UTF-8 character.
1420    <pre>
1421      PCRE_ERROR_PARTIAL        (-12)
1422    </pre>
1423    The subject string did not match, but it did match partially. See the
1424    <a href="pcrepartial.html"><b>pcrepartial</b></a>
1425    documentation for details of partial matching.
1426    <pre>
1427      PCRE_ERROR_BADPARTIAL     (-13)
1428    </pre>
1429    The PCRE_PARTIAL option was used with a compiled pattern containing items that
1430    are not supported for partial matching. See the
1431    <a href="pcrepartial.html"><b>pcrepartial</b></a>
1432    documentation for details of partial matching.
1433    <pre>
1434      PCRE_ERROR_INTERNAL       (-14)
1435    </pre>
1436    An unexpected internal error has occurred. This error could be caused by a bug
1437    in PCRE or by overwriting of the compiled pattern.
1438    <pre>
1439      PCRE_ERROR_BADCOUNT       (-15)
1440    </pre>
1441    This error is given if the value of the <i>ovecsize</i> argument is negative.
1442    <pre>
1443      PCRE_ERROR_RECURSIONLIMIT (-21)
1444    </pre>
1445    The internal recursion limit, as specified by the <i>match_limit_recursion</i>
1446  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
1447  description above.  description above.
 </P>  
 <P>  
1448  <pre>  <pre>
1449    PCRE_ERROR_CALLOUT        (-9)    PCRE_ERROR_NULLWSLIMIT    (-22)
1450  </PRE>  </pre>
1451    When a group that can match an empty substring is repeated with an unbounded
1452    upper limit, the subject position at the start of the group must be remembered,
1453    so that a test for an empty string can be made when the end of the group is
1454    reached. Some workspace is required for this; if it runs out, this error is
1455    given.
1456    <pre>
1457      PCRE_ERROR_BADNEWLINE     (-23)
1458    </pre>
1459    An invalid combination of PCRE_NEWLINE_<i>xxx</i> options was given.
1460  </P>  </P>
1461  <P>  <P>
1462  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.  
1463  </P>  </P>
1464  <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>
1465  <P>  <P>
1466  <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>
1467  <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 1483  Captured substrings can be accessed dire
1483  <b>pcre_get_substring_list()</b> are provided for extracting captured substrings  <b>pcre_get_substring_list()</b> are provided for extracting captured substrings
1484  as new, separate, zero-terminated strings. These functions identify substrings  as new, separate, zero-terminated strings. These functions identify substrings
1485  by number. The next section describes functions for extracting named  by number. The next section describes functions for extracting named
1486  substrings. A substring that contains a binary zero is correctly extracted and  substrings.
1487  has a further zero added on the end, but the result is not, of course,  </P>
1488  a C string.  <P>
1489    A substring that contains a binary zero is correctly extracted and has a
1490    further zero added on the end, but the result is not, of course, a C string.
1491    However, you can process such a string by referring to the length that is
1492    returned by <b>pcre_copy_substring()</b> and <b>pcre_get_substring()</b>.
1493    Unfortunately, the interface to <b>pcre_get_substring_list()</b> is not adequate
1494    for handling strings containing binary zeros, because the end of the final
1495    string is not independently indicated.
1496  </P>  </P>
1497  <P>  <P>
1498  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:
1499  <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,
1500  <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
1501  <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
1502  captured by the match, including the substring that matched the entire regular  captured by the match, including the substring that matched the entire regular
1503  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
1504  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
1505  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
1506  the vector divided by three.  number of elements in the vector divided by three.
1507  </P>  </P>
1508  <P>  <P>
1509  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>
1510  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
1511  value of zero extracts the substring that matched the entire pattern, while  value of zero extracts the substring that matched the entire pattern, whereas
1512  higher values extract the captured substrings. For <b>pcre_copy_substring()</b>,  higher values extract the captured substrings. For <b>pcre_copy_substring()</b>,
1513  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
1514  <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
1515  obtained via <b>pcre_malloc</b>, and its address is returned via  obtained via <b>pcre_malloc</b>, and its address is returned via
1516  <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
1517  including the terminating zero, or one of  including the terminating zero, or one of these error codes:
 </P>  
 <P>  
1518  <pre>  <pre>
1519    PCRE_ERROR_NOMEMORY       (-6)    PCRE_ERROR_NOMEMORY       (-6)
1520  </PRE>  </pre>
 </P>  
 <P>  
1521  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
1522  memory failed for <b>pcre_get_substring()</b>.  memory failed for <b>pcre_get_substring()</b>.
 </P>  
 <P>  
1523  <pre>  <pre>
1524    PCRE_ERROR_NOSUBSTRING    (-7)    PCRE_ERROR_NOSUBSTRING    (-7)
1525  </PRE>  </pre>
 </P>  
 <P>  
1526  There is no substring whose number is <i>stringnumber</i>.  There is no substring whose number is <i>stringnumber</i>.
1527  </P>  </P>
1528  <P>  <P>
1529  The <b>pcre_get_substring_list()</b> function extracts all available substrings  The <b>pcre_get_substring_list()</b> function extracts all available substrings
1530  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
1531  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
1532  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
1533  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
1534  function is zero if all went well, or  function is zero if all went well, or the error code
 </P>  
 <P>  
1535  <pre>  <pre>
1536    PCRE_ERROR_NOMEMORY       (-6)    PCRE_ERROR_NOMEMORY       (-6)
1537  </PRE>  </pre>
 </P>  
 <P>  
1538  if the attempt to get the memory block failed.  if the attempt to get the memory block failed.
1539  </P>  </P>
1540  <P>  <P>
# Line 1202  a previous call of <b>pcre_get_substring Line 1552  a previous call of <b>pcre_get_substring
1552  <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
1553  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
1554  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
1555  linked via a special interface to another programming language which cannot use  linked via a special interface to another programming language that cannot use
1556  <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
1557  provided.  provided.
1558  </P>  </P>
1559  <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>
1560    <P>
1561    <b>int pcre_get_stringnumber(const pcre *<i>code</i>,</b>
1562    <b>const char *<i>name</i>);</b>
1563    </P>
1564  <P>  <P>
1565  <b>int pcre_copy_named_substring(const pcre *<i>code</i>,</b>  <b>int pcre_copy_named_substring(const pcre *<i>code</i>,</b>
1566  <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 1568  provided.
1568  <b>char *<i>buffer</i>, int <i>buffersize</i>);</b>  <b>char *<i>buffer</i>, int <i>buffersize</i>);</b>
1569  </P>  </P>
1570  <P>  <P>
 <b>int pcre_get_stringnumber(const pcre *<i>code</i>,</b>  
 <b>const char *<i>name</i>);</b>  
 </P>  
 <P>  
1571  <b>int pcre_get_named_substring(const pcre *<i>code</i>,</b>  <b>int pcre_get_named_substring(const pcre *<i>code</i>,</b>
1572  <b>const char *<i>subject</i>, int *<i>ovector</i>,</b>  <b>const char *<i>subject</i>, int *<i>ovector</i>,</b>
1573  <b>int <i>stringcount</i>, const char *<i>stringname</i>,</b>  <b>int <i>stringcount</i>, const char *<i>stringname</i>,</b>
1574  <b>const char **<i>stringptr</i>);</b>  <b>const char **<i>stringptr</i>);</b>
1575  </P>  </P>
1576  <P>  <P>
1577  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.
1578  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>  
1579  <pre>  <pre>
1580    ab(?&#60;xxx&#62;\d+)...    (a+)b(?&#60;xxx&#62;\d+)...
1581  </PRE>  </pre>
1582  </P>  the number of the subpattern called "xxx" is 2. If the name is known to be
1583  <P>  unique (PCRE_DUPNAMES was not set), you can find the number from the name by
1584  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
1585  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
1586  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
1587  whole job.  that name.
1588  </P>  </P>
1589  <P>  <P>
1590  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
1591  <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
1592  extract by number, and so are not re-described here. There are just two  two functions that do the whole job.
1593  differences.  </P>
1594    <P>
1595    Most of the arguments of <b>pcre_copy_named_substring()</b> and
1596    <b>pcre_get_named_substring()</b> are the same as those for the similarly named
1597    functions that extract by number. As these are described in the previous
1598    section, they are not re-described here. There are just two differences:
1599  </P>  </P>
1600  <P>  <P>
1601  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 1256  These functions call <b>pcre_get_stringn Line 1608  These functions call <b>pcre_get_stringn
1608  then call <i>pcre_copy_substring()</i> or <i>pcre_get_substring()</i>, as  then call <i>pcre_copy_substring()</i> or <i>pcre_get_substring()</i>, as
1609  appropriate.  appropriate.
1610  </P>  </P>
1611    <br><a name="SEC17" href="#TOC1">DUPLICATE SUBPATTERN NAMES</a><br>
1612    <P>
1613    <b>int pcre_get_stringtable_entries(const pcre *<i>code</i>,</b>
1614    <b>const char *<i>name</i>, char **<i>first</i>, char **<i>last</i>);</b>
1615    </P>
1616    <P>
1617    When a pattern is compiled with the PCRE_DUPNAMES option, names for subpatterns
1618    are not required to be unique. Normally, patterns with duplicate names are such
1619    that in any one match, only one of the named subpatterns participates. An
1620    example is shown in the
1621    <a href="pcrepattern.html"><b>pcrepattern</b></a>
1622    documentation. When duplicates are present, <b>pcre_copy_named_substring()</b>
1623    and <b>pcre_get_named_substring()</b> return the first substring corresponding
1624    to the given name that is set. If none are set, an empty string is returned.
1625    The <b>pcre_get_stringnumber()</b> function returns one of the numbers that are
1626    associated with the name, but it is not defined which it is.
1627    <br>
1628    <br>
1629    If you want to get full details of all captured substrings for a given name,
1630    you must use the <b>pcre_get_stringtable_entries()</b> function. The first
1631    argument is the compiled pattern, and the second is the name. The third and
1632    fourth are pointers to variables which are updated by the function. After it
1633    has run, they point to the first and last entries in the name-to-number table
1634    for the given name. The function itself returns the length of each entry, or
1635    PCRE_ERROR_NOSUBSTRING (-7) if there are none. The format of the table is
1636    described above in the section entitled <i>Information about a pattern</i>.
1637    Given all the relevant entries for the name, you can extract each of their
1638    numbers, and hence the captured data, if any.
1639    </P>
1640    <br><a name="SEC18" href="#TOC1">FINDING ALL POSSIBLE MATCHES</a><br>
1641    <P>
1642    The traditional matching function uses a similar algorithm to Perl, which stops
1643    when it finds the first match, starting at a given point in the subject. If you
1644    want to find all possible matches, or the longest possible match, consider
1645    using the alternative matching function (see below) instead. If you cannot use
1646    the alternative function, but still need to find all possible matches, you
1647    can kludge it up by making use of the callout facility, which is described in
1648    the
1649    <a href="pcrecallout.html"><b>pcrecallout</b></a>
1650    documentation.
1651    </P>
1652    <P>
1653    What you have to do is to insert a callout right at the end of the pattern.
1654    When your callout function is called, extract and save the current matched
1655    substring. Then return 1, which forces <b>pcre_exec()</b> to backtrack and try
1656    other alternatives. Ultimately, when it runs out of matches, <b>pcre_exec()</b>
1657    will yield PCRE_ERROR_NOMATCH.
1658    <a name="dfamatch"></a></P>
1659    <br><a name="SEC19" href="#TOC1">MATCHING A PATTERN: THE ALTERNATIVE FUNCTION</a><br>
1660    <P>
1661    <b>int pcre_dfa_exec(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>
1662    <b>const char *<i>subject</i>, int <i>length</i>, int <i>startoffset</i>,</b>
1663    <b>int <i>options</i>, int *<i>ovector</i>, int <i>ovecsize</i>,</b>
1664    <b>int *<i>workspace</i>, int <i>wscount</i>);</b>
1665    </P>
1666    <P>
1667    The function <b>pcre_dfa_exec()</b> is called to match a subject string against
1668    a compiled pattern, using a matching algorithm that scans the subject string
1669    just once, and does not backtrack. This has different characteristics to the
1670    normal algorithm, and is not compatible with Perl. Some of the features of PCRE
1671    patterns are not supported. Nevertheless, there are times when this kind of
1672    matching can be useful. For a discussion of the two matching algorithms, see
1673    the
1674    <a href="pcrematching.html"><b>pcrematching</b></a>
1675    documentation.
1676    </P>
1677    <P>
1678    The arguments for the <b>pcre_dfa_exec()</b> function are the same as for
1679    <b>pcre_exec()</b>, plus two extras. The <i>ovector</i> argument is used in a
1680    different way, and this is described below. The other common arguments are used
1681    in the same way as for <b>pcre_exec()</b>, so their description is not repeated
1682    here.
1683    </P>
1684    <P>
1685    The two additional arguments provide workspace for the function. The workspace
1686    vector should contain at least 20 elements. It is used for keeping track of
1687    multiple paths through the pattern tree. More workspace will be needed for
1688    patterns and subjects where there are a lot of potential matches.
1689    </P>
1690    <P>
1691    Here is an example of a simple call to <b>pcre_dfa_exec()</b>:
1692    <pre>
1693      int rc;
1694      int ovector[10];
1695      int wspace[20];
1696      rc = pcre_dfa_exec(
1697        re,             /* result of pcre_compile() */
1698        NULL,           /* we didn't study the pattern */
1699        "some string",  /* the subject string */
1700        11,             /* the length of the subject string */
1701        0,              /* start at offset 0 in the subject */
1702        0,              /* default options */
1703        ovector,        /* vector of integers for substring information */
1704        10,             /* number of elements (NOT size in bytes) */
1705        wspace,         /* working space vector */
1706        20);            /* number of elements (NOT size in bytes) */
1707    </PRE>
1708    </P>
1709    <br><b>
1710    Option bits for <b>pcre_dfa_exec()</b>
1711    </b><br>
1712    <P>
1713    The unused bits of the <i>options</i> argument for <b>pcre_dfa_exec()</b> must be
1714    zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_<i>xxx</i>,
1715    PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NO_UTF8_CHECK, PCRE_PARTIAL,
1716    PCRE_DFA_SHORTEST, and PCRE_DFA_RESTART. All but the last three of these are
1717    the same as for <b>pcre_exec()</b>, so their description is not repeated here.
1718    <pre>
1719      PCRE_PARTIAL
1720    </pre>
1721    This has the same general effect as it does for <b>pcre_exec()</b>, but the
1722    details are slightly different. When PCRE_PARTIAL is set for
1723    <b>pcre_dfa_exec()</b>, the return code PCRE_ERROR_NOMATCH is converted into
1724    PCRE_ERROR_PARTIAL if the end of the subject is reached, there have been no
1725    complete matches, but there is still at least one matching possibility. The
1726    portion of the string that provided the partial match is set as the first
1727    matching string.
1728    <pre>
1729      PCRE_DFA_SHORTEST
1730    </pre>
1731    Setting the PCRE_DFA_SHORTEST option causes the matching algorithm to stop as
1732    soon as it has found one match. Because of the way the alternative algorithm
1733    works, this is necessarily the shortest possible match at the first possible
1734    matching point in the subject string.
1735    <pre>
1736      PCRE_DFA_RESTART
1737    </pre>
1738    When <b>pcre_dfa_exec()</b> is called with the PCRE_PARTIAL option, and returns
1739    a partial match, it is possible to call it again, with additional subject
1740    characters, and have it continue with the same match. The PCRE_DFA_RESTART
1741    option requests this action; when it is set, the <i>workspace</i> and
1742    <i>wscount</i> options must reference the same vector as before because data
1743    about the match so far is left in them after a partial match. There is more
1744    discussion of this facility in the
1745    <a href="pcrepartial.html"><b>pcrepartial</b></a>
1746    documentation.
1747    </P>
1748    <br><b>
1749    Successful returns from <b>pcre_dfa_exec()</b>
1750    </b><br>
1751    <P>
1752    When <b>pcre_dfa_exec()</b> succeeds, it may have matched more than one
1753    substring in the subject. Note, however, that all the matches from one run of
1754    the function start at the same point in the subject. The shorter matches are
1755    all initial substrings of the longer matches. For example, if the pattern
1756    <pre>
1757      &#60;.*&#62;
1758    </pre>
1759    is matched against the string
1760    <pre>
1761      This is &#60;something&#62; &#60;something else&#62; &#60;something further&#62; no more
1762    </pre>
1763    the three matched strings are
1764    <pre>
1765      &#60;something&#62;
1766      &#60;something&#62; &#60;something else&#62;
1767      &#60;something&#62; &#60;something else&#62; &#60;something further&#62;
1768    </pre>
1769    On success, the yield of the function is a number greater than zero, which is
1770    the number of matched substrings. The substrings themselves are returned in
1771    <i>ovector</i>. Each string uses two elements; the first is the offset to the
1772    start, and the second is the offset to the end. In fact, all the strings have
1773    the same start offset. (Space could have been saved by giving this only once,
1774    but it was decided to retain some compatibility with the way <b>pcre_exec()</b>
1775    returns data, even though the meaning of the strings is different.)
1776    </P>
1777    <P>
1778    The strings are returned in reverse order of length; that is, the longest
1779    matching string is given first. If there were too many matches to fit into
1780    <i>ovector</i>, the yield of the function is zero, and the vector is filled with
1781    the longest matches.
1782    </P>
1783    <br><b>
1784    Error returns from <b>pcre_dfa_exec()</b>
1785    </b><br>
1786    <P>
1787    The <b>pcre_dfa_exec()</b> function returns a negative number when it fails.
1788    Many of the errors are the same as for <b>pcre_exec()</b>, and these are
1789    described
1790    <a href="#errorlist">above.</a>
1791    There are in addition the following errors that are specific to
1792    <b>pcre_dfa_exec()</b>:
1793    <pre>
1794      PCRE_ERROR_DFA_UITEM      (-16)
1795    </pre>
1796    This return is given if <b>pcre_dfa_exec()</b> encounters an item in the pattern
1797    that it does not support, for instance, the use of \C or a back reference.
1798    <pre>
1799      PCRE_ERROR_DFA_UCOND      (-17)
1800    </pre>
1801    This return is given if <b>pcre_dfa_exec()</b> encounters a condition item that
1802    uses a back reference for the condition, or a test for recursion in a specific
1803    group. These are not supported.
1804    <pre>
1805      PCRE_ERROR_DFA_UMLIMIT    (-18)
1806    </pre>
1807    This return is given if <b>pcre_dfa_exec()</b> is called with an <i>extra</i>
1808    block that contains a setting of the <i>match_limit</i> field. This is not
1809    supported (it is meaningless).
1810    <pre>
1811      PCRE_ERROR_DFA_WSSIZE     (-19)
1812    </pre>
1813    This return is given if <b>pcre_dfa_exec()</b> runs out of space in the
1814    <i>workspace</i> vector.
1815    <pre>
1816      PCRE_ERROR_DFA_RECURSE    (-20)
1817    </pre>
1818    When a recursive subpattern is processed, the matching function calls itself
1819    recursively, using private vectors for <i>ovector</i> and <i>workspace</i>. This
1820    error is given if the output vector is not large enough. This should be
1821    extremely rare, as a vector of size 1000 is used.
1822    </P>
1823    <br><a name="SEC20" href="#TOC1">SEE ALSO</a><br>
1824    <P>
1825    <b>pcrebuild</b>(3), <b>pcrecallout</b>(3), <b>pcrecpp(3)</b>(3),
1826    <b>pcrematching</b>(3), <b>pcrepartial</b>(3), <b>pcreposix</b>(3),
1827    <b>pcreprecompile</b>(3), <b>pcresample</b>(3), <b>pcrestack</b>(3).
1828    </P>
1829  <P>  <P>
1830  Last updated: 03 February 2003  Last updated: 30 November 2006
1831  <br>  <br>
1832  Copyright &copy; 1997-2003 University of Cambridge.  Copyright &copy; 1997-2006 University of Cambridge.
1833    <p>
1834    Return to the <a href="index.html">PCRE index page</a>.
1835    </p>

Legend:
Removed from v.65  
changed lines
  Added in v.96

  ViewVC Help
Powered by ViewVC 1.1.5