/[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 243 by ph10, Thu Sep 13 09:28:14 2007 UTC
# Line 3  Line 3 
3  <title>pcreapi specification</title>  <title>pcreapi specification</title>
4  </head>  </head>
5  <body bgcolor="#FFFFFF" text="#00005A" link="#0066FF" alink="#3399FF" vlink="#2222BB">  <body bgcolor="#FFFFFF" text="#00005A" link="#0066FF" alink="#3399FF" vlink="#2222BB">
6  This HTML document has been generated automatically from the original man page.  <h1>pcreapi man page</h1>
7  If there is any nonsense in it, please consult the man page, in case the  <p>
8  conversion went wrong.<br>  Return to the <a href="index.html">PCRE index page</a>.
9    </p>
10    <p>
11    This page is part of the PCRE HTML documentation. It was generated automatically
12    from the original man page. If there is any nonsense in it, please consult the
13    man page, in case the conversion went wrong.
14    <br>
15  <ul>  <ul>
16  <li><a name="TOC1" href="#SEC1">SYNOPSIS OF PCRE API</a>  <li><a name="TOC1" href="#SEC1">PCRE NATIVE API</a>
17  <li><a name="TOC2" href="#SEC2">PCRE API</a>  <li><a name="TOC2" href="#SEC2">PCRE API OVERVIEW</a>
18  <li><a name="TOC3" href="#SEC3">MULTITHREADING</a>  <li><a name="TOC3" href="#SEC3">NEWLINES</a>
19  <li><a name="TOC4" href="#SEC4">CHECKING BUILD-TIME OPTIONS</a>  <li><a name="TOC4" href="#SEC4">MULTITHREADING</a>
20  <li><a name="TOC5" href="#SEC5">COMPILING A PATTERN</a>  <li><a name="TOC5" href="#SEC5">SAVING PRECOMPILED PATTERNS FOR LATER USE</a>
21  <li><a name="TOC6" href="#SEC6">STUDYING A PATTERN</a>  <li><a name="TOC6" href="#SEC6">CHECKING BUILD-TIME OPTIONS</a>
22  <li><a name="TOC7" href="#SEC7">LOCALE SUPPORT</a>  <li><a name="TOC7" href="#SEC7">COMPILING A PATTERN</a>
23  <li><a name="TOC8" href="#SEC8">INFORMATION ABOUT A PATTERN</a>  <li><a name="TOC8" href="#SEC8">COMPILATION ERROR CODES</a>
24  <li><a name="TOC9" href="#SEC9">OBSOLETE INFO FUNCTION</a>  <li><a name="TOC9" href="#SEC9">STUDYING A PATTERN</a>
25  <li><a name="TOC10" href="#SEC10">MATCHING A PATTERN</a>  <li><a name="TOC10" href="#SEC10">LOCALE SUPPORT</a>
26  <li><a name="TOC11" href="#SEC11">EXTRACTING CAPTURED SUBSTRINGS BY NUMBER</a>  <li><a name="TOC11" href="#SEC11">INFORMATION ABOUT A PATTERN</a>
27  <li><a name="TOC12" href="#SEC12">EXTRACTING CAPTURED SUBSTRINGS BY NAME</a>  <li><a name="TOC12" href="#SEC12">OBSOLETE INFO FUNCTION</a>
28    <li><a name="TOC13" href="#SEC13">REFERENCE COUNTS</a>
29    <li><a name="TOC14" href="#SEC14">MATCHING A PATTERN: THE TRADITIONAL FUNCTION</a>
30    <li><a name="TOC15" href="#SEC15">EXTRACTING CAPTURED SUBSTRINGS BY NUMBER</a>
31    <li><a name="TOC16" href="#SEC16">EXTRACTING CAPTURED SUBSTRINGS BY NAME</a>
32    <li><a name="TOC17" href="#SEC17">DUPLICATE SUBPATTERN NAMES</a>
33    <li><a name="TOC18" href="#SEC18">FINDING ALL POSSIBLE MATCHES</a>
34    <li><a name="TOC19" href="#SEC19">MATCHING A PATTERN: THE ALTERNATIVE FUNCTION</a>
35    <li><a name="TOC20" href="#SEC20">SEE ALSO</a>
36    <li><a name="TOC21" href="#SEC21">AUTHOR</a>
37    <li><a name="TOC22" href="#SEC22">REVISION</a>
38  </ul>  </ul>
39  <br><a name="SEC1" href="#TOC1">SYNOPSIS OF PCRE API</a><br>  <br><a name="SEC1" href="#TOC1">PCRE NATIVE API</a><br>
40  <P>  <P>
41  <b>#include &#60;pcre.h&#62;</b>  <b>#include &#60;pcre.h&#62;</b>
42  </P>  </P>
# Line 30  conversion went wrong.<br> Line 46  conversion went wrong.<br>
46  <b>const unsigned char *<i>tableptr</i>);</b>  <b>const unsigned char *<i>tableptr</i>);</b>
47  </P>  </P>
48  <P>  <P>
49    <b>pcre *pcre_compile2(const char *<i>pattern</i>, int <i>options</i>,</b>
50    <b>int *<i>errorcodeptr</i>,</b>
51    <b>const char **<i>errptr</i>, int *<i>erroffset</i>,</b>
52    <b>const unsigned char *<i>tableptr</i>);</b>
53    </P>
54    <P>
55  <b>pcre_extra *pcre_study(const pcre *<i>code</i>, int <i>options</i>,</b>  <b>pcre_extra *pcre_study(const pcre *<i>code</i>, int <i>options</i>,</b>
56  <b>const char **<i>errptr</i>);</b>  <b>const char **<i>errptr</i>);</b>
57  </P>  </P>
# Line 39  conversion went wrong.<br> Line 61  conversion went wrong.<br>
61  <b>int <i>options</i>, int *<i>ovector</i>, int <i>ovecsize</i>);</b>  <b>int <i>options</i>, int *<i>ovector</i>, int <i>ovecsize</i>);</b>
62  </P>  </P>
63  <P>  <P>
64    <b>int pcre_dfa_exec(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>
65    <b>const char *<i>subject</i>, int <i>length</i>, int <i>startoffset</i>,</b>
66    <b>int <i>options</i>, int *<i>ovector</i>, int <i>ovecsize</i>,</b>
67    <b>int *<i>workspace</i>, int <i>wscount</i>);</b>
68    </P>
69    <P>
70  <b>int pcre_copy_named_substring(const pcre *<i>code</i>,</b>  <b>int pcre_copy_named_substring(const pcre *<i>code</i>,</b>
71  <b>const char *<i>subject</i>, int *<i>ovector</i>,</b>  <b>const char *<i>subject</i>, int *<i>ovector</i>,</b>
72  <b>int <i>stringcount</i>, const char *<i>stringname</i>,</b>  <b>int <i>stringcount</i>, const char *<i>stringname</i>,</b>
# Line 60  conversion went wrong.<br> Line 88  conversion went wrong.<br>
88  <b>const char *<i>name</i>);</b>  <b>const char *<i>name</i>);</b>
89  </P>  </P>
90  <P>  <P>
91    <b>int pcre_get_stringtable_entries(const pcre *<i>code</i>,</b>
92    <b>const char *<i>name</i>, char **<i>first</i>, char **<i>last</i>);</b>
93    </P>
94    <P>
95  <b>int pcre_get_substring(const char *<i>subject</i>, int *<i>ovector</i>,</b>  <b>int pcre_get_substring(const char *<i>subject</i>, int *<i>ovector</i>,</b>
96  <b>int <i>stringcount</i>, int <i>stringnumber</i>,</b>  <b>int <i>stringcount</i>, int <i>stringnumber</i>,</b>
97  <b>const char **<i>stringptr</i>);</b>  <b>const char **<i>stringptr</i>);</b>
# Line 86  conversion went wrong.<br> Line 118  conversion went wrong.<br>
118  <b>*<i>firstcharptr</i>);</b>  <b>*<i>firstcharptr</i>);</b>
119  </P>  </P>
120  <P>  <P>
121    <b>int pcre_refcount(pcre *<i>code</i>, int <i>adjust</i>);</b>
122    </P>
123    <P>
124  <b>int pcre_config(int <i>what</i>, void *<i>where</i>);</b>  <b>int pcre_config(int <i>what</i>, void *<i>where</i>);</b>
125  </P>  </P>
126  <P>  <P>
# Line 98  conversion went wrong.<br> Line 133  conversion went wrong.<br>
133  <b>void (*pcre_free)(void *);</b>  <b>void (*pcre_free)(void *);</b>
134  </P>  </P>
135  <P>  <P>
136  <b>int (*pcre_callout)(pcre_callout_block *);</b>  <b>void *(*pcre_stack_malloc)(size_t);</b>
137  </P>  </P>
 <br><a name="SEC2" href="#TOC1">PCRE API</a><br>  
138  <P>  <P>
139  PCRE has its own native API, which is described in this document. There is also  <b>void (*pcre_stack_free)(void *);</b>
 a set of wrapper functions that correspond to the POSIX regular expression API.  
 These are described in the <b>pcreposix</b> documentation.  
140  </P>  </P>
141  <P>  <P>
142  The native API function prototypes are defined in the header file <b>pcre.h</b>,  <b>int (*pcre_callout)(pcre_callout_block *);</b>
 and on Unix systems the library itself is called <b>libpcre.a</b>, so can be  
 accessed by adding <b>-lpcre</b> to the command for linking an application which  
 calls it. The header file defines the macros PCRE_MAJOR and PCRE_MINOR to  
 contain the major and minor release numbers for the library. Applications can  
 use these to include support for different releases.  
143  </P>  </P>
144    <br><a name="SEC2" href="#TOC1">PCRE API OVERVIEW</a><br>
145  <P>  <P>
146  The functions <b>pcre_compile()</b>, <b>pcre_study()</b>, and <b>pcre_exec()</b>  PCRE has its own native API, which is described in this document. There are
147  are used for compiling and matching regular expressions. A sample program that  also some wrapper functions that correspond to the POSIX regular expression
148  demonstrates the simplest way of using them is given in the file  API. These are described in the
149  <i>pcredemo.c</i>. The <b>pcresample</b> documentation describes how to run it.  <a href="pcreposix.html"><b>pcreposix</b></a>
150    documentation. Both of these APIs define a set of C function calls. A C++
151    wrapper is distributed with PCRE. It is documented in the
152    <a href="pcrecpp.html"><b>pcrecpp</b></a>
153    page.
154  </P>  </P>
155  <P>  <P>
156  There are convenience functions for extracting captured substrings from a  The native API C function prototypes are defined in the header file
157  matched subject string. They are:  <b>pcre.h</b>, and on Unix systems the library itself is called <b>libpcre</b>.
158    It can normally be accessed by adding <b>-lpcre</b> to the command for linking
159    an application that uses PCRE. The header file defines the macros PCRE_MAJOR
160    and PCRE_MINOR to contain the major and minor release numbers for the library.
161    Applications can use these to include support for different releases of PCRE.
162    </P>
163    <P>
164    The functions <b>pcre_compile()</b>, <b>pcre_compile2()</b>, <b>pcre_study()</b>,
165    and <b>pcre_exec()</b> are used for compiling and matching regular expressions
166    in a Perl-compatible manner. A sample program that demonstrates the simplest
167    way of using them is provided in the file called <i>pcredemo.c</i> in the source
168    distribution. The
169    <a href="pcresample.html"><b>pcresample</b></a>
170    documentation describes how to run it.
171    </P>
172    <P>
173    A second matching function, <b>pcre_dfa_exec()</b>, which is not
174    Perl-compatible, is also provided. This uses a different algorithm for the
175    matching. The alternative algorithm finds all possible matches (at a given
176    point in the subject), and scans the subject just once. However, this algorithm
177    does not return captured substrings. A description of the two matching
178    algorithms and their advantages and disadvantages is given in the
179    <a href="pcrematching.html"><b>pcrematching</b></a>
180    documentation.
181  </P>  </P>
182  <P>  <P>
183    In addition to the main compiling and matching functions, there are convenience
184    functions for extracting captured substrings from a subject string that is
185    matched by <b>pcre_exec()</b>. They are:
186  <pre>  <pre>
187    <b>pcre_copy_substring()</b>    <b>pcre_copy_substring()</b>
188    <b>pcre_copy_named_substring()</b>    <b>pcre_copy_named_substring()</b>
189    <b>pcre_get_substring()</b>    <b>pcre_get_substring()</b>
190    <b>pcre_get_named_substring()</b>    <b>pcre_get_named_substring()</b>
191    <b>pcre_get_substring_list()</b>    <b>pcre_get_substring_list()</b>
192  </PRE>    <b>pcre_get_stringnumber()</b>
193  </P>    <b>pcre_get_stringtable_entries()</b>
194  <P>  </pre>
195  <b>pcre_free_substring()</b> and <b>pcre_free_substring_list()</b> are also  <b>pcre_free_substring()</b> and <b>pcre_free_substring_list()</b> are also
196  provided, to free the memory used for extracted strings.  provided, to free the memory used for extracted strings.
197  </P>  </P>
198  <P>  <P>
199  The function <b>pcre_maketables()</b> is used (optionally) to build a set of  The function <b>pcre_maketables()</b> is used to build a set of character tables
200  character tables in the current locale for passing to <b>pcre_compile()</b>.  in the current locale for passing to <b>pcre_compile()</b>, <b>pcre_exec()</b>,
201    or <b>pcre_dfa_exec()</b>. This is an optional facility that is provided for
202    specialist use. Most commonly, no special tables are passed, in which case
203    internal tables that are generated when PCRE is built are used.
204  </P>  </P>
205  <P>  <P>
206  The function <b>pcre_fullinfo()</b> is used to find out information about a  The function <b>pcre_fullinfo()</b> is used to find out information about a
207  compiled pattern; <b>pcre_info()</b> is an obsolete version which returns only  compiled pattern; <b>pcre_info()</b> is an obsolete version that returns only
208  some of the available information, but is retained for backwards compatibility.  some of the available information, but is retained for backwards compatibility.
209  The function <b>pcre_version()</b> returns a pointer to a string containing the  The function <b>pcre_version()</b> returns a pointer to a string containing the
210  version of PCRE and its date of release.  version of PCRE and its date of release.
211  </P>  </P>
212  <P>  <P>
213    The function <b>pcre_refcount()</b> maintains a reference count in a data block
214    containing a compiled pattern. This is provided for the benefit of
215    object-oriented applications.
216    </P>
217    <P>
218  The global variables <b>pcre_malloc</b> and <b>pcre_free</b> initially contain  The global variables <b>pcre_malloc</b> and <b>pcre_free</b> initially contain
219  the entry points of the standard <b>malloc()</b> and <b>free()</b> functions  the entry points of the standard <b>malloc()</b> and <b>free()</b> functions,
220  respectively. PCRE calls the memory management functions via these variables,  respectively. PCRE calls the memory management functions via these variables,
221  so a calling program can replace them if it wishes to intercept the calls. This  so a calling program can replace them if it wishes to intercept the calls. This
222  should be done before calling any PCRE functions.  should be done before calling any PCRE functions.
223  </P>  </P>
224  <P>  <P>
225    The global variables <b>pcre_stack_malloc</b> and <b>pcre_stack_free</b> are also
226    indirections to memory management functions. These special functions are used
227    only when PCRE is compiled to use the heap for remembering data, instead of
228    recursive function calls, when running the <b>pcre_exec()</b> function. See the
229    <a href="pcrebuild.html"><b>pcrebuild</b></a>
230    documentation for details of how to do this. It is a non-standard way of
231    building PCRE, for use in environments that have limited stacks. Because of the
232    greater use of memory management, it runs more slowly. Separate functions are
233    provided so that special-purpose external code can be used for this case. When
234    used, these functions are always called in a stack-like manner (last obtained,
235    first freed), and always for memory blocks of the same size. There is a
236    discussion about PCRE's stack usage in the
237    <a href="pcrestack.html"><b>pcrestack</b></a>
238    documentation.
239    </P>
240    <P>
241  The global variable <b>pcre_callout</b> initially contains NULL. It can be set  The global variable <b>pcre_callout</b> initially contains NULL. It can be set
242  by the caller to a "callout" function, which PCRE will then call at specified  by the caller to a "callout" function, which PCRE will then call at specified
243  points during a matching operation. Details are given in the <b>pcrecallout</b>  points during a matching operation. Details are given in the
244    <a href="pcrecallout.html"><b>pcrecallout</b></a>
245  documentation.  documentation.
246    <a name="newlines"></a></P>
247    <br><a name="SEC3" href="#TOC1">NEWLINES</a><br>
248    <P>
249    PCRE supports five different conventions for indicating line breaks in
250    strings: a single CR (carriage return) character, a single LF (linefeed)
251    character, the two-character sequence CRLF, any of the three preceding, or any
252    Unicode newline sequence. The Unicode newline sequences are the three just
253    mentioned, plus the single characters VT (vertical tab, U+000B), FF (formfeed,
254    U+000C), NEL (next line, U+0085), LS (line separator, U+2028), and PS
255    (paragraph separator, U+2029).
256    </P>
257    <P>
258    Each of the first three conventions is used by at least one operating system as
259    its standard newline sequence. When PCRE is built, a default can be specified.
260    The default default is LF, which is the Unix standard. When PCRE is run, the
261    default can be overridden, either when a pattern is compiled, or when it is
262    matched.
263    </P>
264    <P>
265    At compile time, the newline convention can be specified by the <i>options</i>
266    argument of <b>pcre_compile()</b>, or it can be specified by special text at the
267    start of the pattern itself; this overrides any other settings. See the
268    <a href="pcrepattern.html"><b>pcrepattern</b></a>
269    page for details of the special character sequences.
270    </P>
271    <P>
272    In the PCRE documentation the word "newline" is used to mean "the character or
273    pair of characters that indicate a line break". The choice of newline
274    convention affects the handling of the dot, circumflex, and dollar
275    metacharacters, the handling of #-comments in /x mode, and, when CRLF is a
276    recognized line ending sequence, the match position advancement for a
277    non-anchored pattern. There is more detail about this in the
278    <a href="#execoptions">section on <b>pcre_exec()</b> options</a>
279    below.
280    </P>
281    <P>
282    The choice of newline convention does not affect the interpretation of
283    the \n or \r escape sequences, nor does it affect what \R matches, which is
284    controlled in a similar way, but by separate options.
285  </P>  </P>
286  <br><a name="SEC3" href="#TOC1">MULTITHREADING</a><br>  <br><a name="SEC4" href="#TOC1">MULTITHREADING</a><br>
287  <P>  <P>
288  The PCRE functions can be used in multi-threading applications, with the  The PCRE functions can be used in multi-threading applications, with the
289  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>,
290  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
291  are shared by all threads.  callout function pointed to by <b>pcre_callout</b>, are shared by all threads.
292  </P>  </P>
293  <P>  <P>
294  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
295  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.
296  </P>  </P>
297  <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>
298    <P>
299    The compiled form of a regular expression can be saved and re-used at a later
300    time, possibly by a different program, and even on a host other than the one on
301    which it was compiled. Details are given in the
302    <a href="pcreprecompile.html"><b>pcreprecompile</b></a>
303    documentation. However, compiling a regular expression with one version of PCRE
304    for use with a different version is not guaranteed to work and may cause
305    crashes.
306    </P>
307    <br><a name="SEC6" href="#TOC1">CHECKING BUILD-TIME OPTIONS</a><br>
308  <P>  <P>
309  <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>
310  </P>  </P>
# Line 186  documentation has more details about the Line 318  documentation has more details about the
318  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
319  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
320  which the information is placed. The following information is available:  which the information is placed. The following information is available:
 </P>  
 <P>  
321  <pre>  <pre>
322    PCRE_CONFIG_UTF8    PCRE_CONFIG_UTF8
323  </PRE>  </pre>
 </P>  
 <P>  
324  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;
325  otherwise it is set to zero.  otherwise it is set to zero.
326  </P>  <pre>
327  <P>    PCRE_CONFIG_UNICODE_PROPERTIES
328    </pre>
329    The output is an integer that is set to one if support for Unicode character
330    properties is available; otherwise it is set to zero.
331  <pre>  <pre>
332    PCRE_CONFIG_NEWLINE    PCRE_CONFIG_NEWLINE
333  </PRE>  </pre>
334  </P>  The output is an integer whose value specifies the default character sequence
335  <P>  that is recognized as meaning "newline". The four values that are supported
336  The output is an integer that is set to the value of the code that is used for  are: 10 for LF, 13 for CR, 3338 for CRLF, -2 for ANYCRLF, and -1 for ANY. The
337  the newline character. It is either linefeed (10) or carriage return (13), and  default should normally be the standard sequence for your operating system.
338  should normally be the standard character for your operating system.  <pre>
339  </P>    PCRE_CONFIG_BSR
340  <P>  </pre>
341    The output is an integer whose value indicates what character sequences the \R
342    escape sequence matches by default. A value of 0 means that \R matches any
343    Unicode line ending sequence; a value of 1 means that \R matches only CR, LF,
344    or CRLF. The default can be overridden when a pattern is compiled or matched.
345  <pre>  <pre>
346    PCRE_CONFIG_LINK_SIZE    PCRE_CONFIG_LINK_SIZE
347  </PRE>  </pre>
 </P>  
 <P>  
348  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
349  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
350  allow larger regular expressions to be compiled, at the expense of slower  allow larger regular expressions to be compiled, at the expense of slower
351  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
352  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>  
353  <pre>  <pre>
354    PCRE_CONFIG_POSIX_MALLOC_THRESHOLD    PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
355  </PRE>  </pre>
 </P>  
 <P>  
356  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
357  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
358  the <b>pcreposix</b> documentation.  the
359  </P>  <a href="pcreposix.html"><b>pcreposix</b></a>
360  <P>  documentation.
361  <pre>  <pre>
362    PCRE_CONFIG_MATCH_LIMIT    PCRE_CONFIG_MATCH_LIMIT
363  </PRE>  </pre>
 </P>  
 <P>  
364  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
365  internal matching function calls in a <b>pcre_exec()</b> execution. Further  internal matching function calls in a <b>pcre_exec()</b> execution. Further
366  details are given with <b>pcre_exec()</b> below.  details are given with <b>pcre_exec()</b> below.
367    <pre>
368      PCRE_CONFIG_MATCH_LIMIT_RECURSION
369    </pre>
370    The output is an integer that gives the default limit for the depth of
371    recursion when calling the internal matching function in a <b>pcre_exec()</b>
372    execution. Further details are given with <b>pcre_exec()</b> below.
373    <pre>
374      PCRE_CONFIG_STACKRECURSE
375    </pre>
376    The output is an integer that is set to one if internal recursion when running
377    <b>pcre_exec()</b> is implemented by recursive function calls that use the stack
378    to remember their state. This is the usual way that PCRE is compiled. The
379    output is zero if PCRE was compiled to use blocks of data on the heap instead
380    of recursive function calls. In this case, <b>pcre_stack_malloc</b> and
381    <b>pcre_stack_free</b> are called to manage memory blocks on the heap, thus
382    avoiding the use of the stack.
383  </P>  </P>
384  <br><a name="SEC5" href="#TOC1">COMPILING A PATTERN</a><br>  <br><a name="SEC7" href="#TOC1">COMPILING A PATTERN</a><br>
385  <P>  <P>
386  <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>
387  <b>const char **<i>errptr</i>, int *<i>erroffset</i>,</b>  <b>const char **<i>errptr</i>, int *<i>erroffset</i>,</b>
388  <b>const unsigned char *<i>tableptr</i>);</b>  <b>const unsigned char *<i>tableptr</i>);</b>
389    <b>pcre *pcre_compile2(const char *<i>pattern</i>, int <i>options</i>,</b>
390    <b>int *<i>errorcodeptr</i>,</b>
391    <b>const char **<i>errptr</i>, int *<i>erroffset</i>,</b>
392    <b>const unsigned char *<i>tableptr</i>);</b>
393    </P>
394    <P>
395    Either of the functions <b>pcre_compile()</b> or <b>pcre_compile2()</b> can be
396    called to compile a pattern into an internal form. The only difference between
397    the two interfaces is that <b>pcre_compile2()</b> has an additional argument,
398    <i>errorcodeptr</i>, via which a numerical error code can be returned.
399  </P>  </P>
400  <P>  <P>
401  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
402  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
403  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
404  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
405  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
406  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.  
407  </P>  </P>
408  <P>  <P>
409  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
410  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
411  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>
412  which is an address (see below).  argument, which is an address (see below).
413  </P>  </P>
414  <P>  <P>
415  The <i>options</i> argument contains independent bits that affect the  The <i>options</i> argument contains various bit settings that affect the
416  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
417  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
418  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
419  in the <b>pcrepattern</b> documentation). For these options, the contents of the  the detailed description in the
420  <i>options</i> argument specifies their initial settings at the start of  <a href="pcrepattern.html"><b>pcrepattern</b></a>
421  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
422    specifies their initial settings at the start of compilation and execution. The
423    PCRE_ANCHORED and PCRE_NEWLINE_<i>xxx</i> options can be set at the time of
424  matching as well as at compile time.  matching as well as at compile time.
425  </P>  </P>
426  <P>  <P>
427  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.
428  Otherwise, if compilation of a pattern fails, <b>pcre_compile()</b> returns  Otherwise, if compilation of a pattern fails, <b>pcre_compile()</b> returns
429  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
430  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
431  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
432    where the error was discovered is placed in the variable pointed to by
433  <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.
434  </P>  </P>
435  <P>  <P>
436  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
437  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
438  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
439  <b>pcre_maketables()</b>. See the section on locale support below.  textual error message. Error codes and messages are listed below.
440  </P>  </P>
441  <P>  <P>
442  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
443    character tables that are built when PCRE is compiled, using the default C
444    locale. Otherwise, <i>tableptr</i> must be an address that is the result of a
445    call to <b>pcre_maketables()</b>. This value is stored with the compiled
446    pattern, and used again by <b>pcre_exec()</b>, unless another table pointer is
447    passed to it. For more discussion, see the section on locale support below.
448  </P>  </P>
449  <P>  <P>
450    This code fragment shows a typical straightforward call to <b>pcre_compile()</b>:
451  <pre>  <pre>
452    pcre *re;    pcre *re;
453    const char *error;    const char *error;
# Line 297  This code fragment shows a typical strai Line 458  This code fragment shows a typical strai
458      &error,           /* for error message */      &error,           /* for error message */
459      &erroffset,       /* for error offset */      &erroffset,       /* for error offset */
460      NULL);            /* use default character tables */      NULL);            /* use default character tables */
461  </PRE>  </pre>
462  </P>  The following names for option bits are defined in the <b>pcre.h</b> header
463  <P>  file:
 The following option bits are defined:  
 </P>  
 <P>  
464  <pre>  <pre>
465    PCRE_ANCHORED    PCRE_ANCHORED
466  </PRE>  </pre>
 </P>  
 <P>  
467  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
468  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
469  being searched (the "subject string"). This effect can also be achieved by  being searched (the "subject string"). This effect can also be achieved by
470  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
471  Perl.  Perl.
472  </P>  <pre>
473  <P>    PCRE_AUTO_CALLOUT
474    </pre>
475    If this bit is set, <b>pcre_compile()</b> automatically inserts callout items,
476    all with number 255, before each pattern item. For discussion of the callout
477    facility, see the
478    <a href="pcrecallout.html"><b>pcrecallout</b></a>
479    documentation.
480    <pre>
481      PCRE_BSR_ANYCRLF
482      PCRE_BSR_UNICODE
483    </pre>
484    These options (which are mutually exclusive) control what the \R escape
485    sequence matches. The choice is either to match only CR, LF, or CRLF, or to
486    match any Unicode newline sequence. The default is specified when PCRE is
487    built. It can be overridden from within the pattern, or by setting an option
488    when a compiled pattern is matched.
489  <pre>  <pre>
490    PCRE_CASELESS    PCRE_CASELESS
491  </PRE>  </pre>
 </P>  
 <P>  
492  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
493  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
494  pattern by a (?i) option setting.  pattern by a (?i) option setting. In UTF-8 mode, PCRE always understands the
495  </P>  concept of case for characters whose values are less than 128, so caseless
496  <P>  matching is always possible. For characters with higher values, the concept of
497    case is supported if PCRE is compiled with Unicode property support, but not
498    otherwise. If you want to use caseless matching for characters 128 and above,
499    you must ensure that PCRE is compiled with Unicode property support as well as
500    with UTF-8 support.
501  <pre>  <pre>
502    PCRE_DOLLAR_ENDONLY    PCRE_DOLLAR_ENDONLY
503  </PRE>  </pre>
 </P>  
 <P>  
504  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
505  end of the subject string. Without this option, a dollar also matches  end of the subject string. Without this option, a dollar also matches
506  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
507  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.
508  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
509  a pattern.  pattern.
 </P>  
 <P>  
510  <pre>  <pre>
511    PCRE_DOTALL    PCRE_DOTALL
512  </PRE>  </pre>
 </P>  
 <P>  
513  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,
514  including newlines. Without it, newlines are excluded. This option is  including those that indicate newline. Without it, a dot does not match when
515  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
516  (?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
517  character, independent of the setting of this option.  negative class such as [^a] always matches newline characters, independent of
518  </P>  the setting of this option.
519  <P>  <pre>
520      PCRE_DUPNAMES
521    </pre>
522    If this bit is set, names used to identify capturing subpatterns need not be
523    unique. This can be helpful for certain types of pattern when it is known that
524    only one instance of the named subpattern can ever be matched. There are more
525    details of named subpatterns below; see also the
526    <a href="pcrepattern.html"><b>pcrepattern</b></a>
527    documentation.
528  <pre>  <pre>
529    PCRE_EXTENDED    PCRE_EXTENDED
530  </PRE>  </pre>
 </P>  
 <P>  
531  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
532  ignored except when escaped or inside a character class. Whitespace does not  ignored except when escaped or inside a character class. Whitespace does not
533  include the VT character (code 11). In addition, characters between an  include the VT character (code 11). In addition, characters between an
534  unescaped # outside a character class and the next newline character,  unescaped # outside a character class and the next newline, inclusive, are also
535  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
536  be changed within a pattern by a (?x) option setting.  pattern by a (?x) option setting.
537  </P>  </P>
538  <P>  <P>
539  This option makes it possible to include comments inside complicated patterns.  This option makes it possible to include comments inside complicated patterns.
540  Note, however, that this applies only to data characters. Whitespace characters  Note, however, that this applies only to data characters. Whitespace characters
541  may never appear within special character sequences in a pattern, for example  may never appear within special character sequences in a pattern, for example
542  within the sequence (?( which introduces a conditional subpattern.  within the sequence (?( which introduces a conditional subpattern.
 </P>  
 <P>  
543  <pre>  <pre>
544    PCRE_EXTRA    PCRE_EXTRA
545  </PRE>  </pre>
 </P>  
 <P>  
546  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
547  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
548  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
549  special meaning causes an error, thus reserving these combinations for future  special meaning causes an error, thus reserving these combinations for future
550  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
551  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
552  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
553  pattern.  this option. It can also be set by a (?X) option setting within a pattern.
554  </P>  <pre>
555  <P>    PCRE_FIRSTLINE
556    </pre>
557    If this option is set, an unanchored pattern is required to match before or at
558    the first newline in the subject string, though the matched text may continue
559    over the newline.
560  <pre>  <pre>
561    PCRE_MULTILINE    PCRE_MULTILINE
562  </PRE>  </pre>
563  </P>  By default, PCRE treats the subject string as consisting of a single line of
564  <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"  
565  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
566  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
567  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 569  Perl.
569  </P>  </P>
570  <P>  <P>
571  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
572  match immediately following or immediately before any newline in the subject  match immediately following or immediately before internal newlines in the
573  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
574  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
575  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
576  occurrences of ^ or $ in a pattern, setting PCRE_MULTILINE has no effect.  occurrences of ^ or $ in a pattern, setting PCRE_MULTILINE has no effect.
577    <pre>
578      PCRE_NEWLINE_CR
579      PCRE_NEWLINE_LF
580      PCRE_NEWLINE_CRLF
581      PCRE_NEWLINE_ANYCRLF
582      PCRE_NEWLINE_ANY
583    </pre>
584    These options override the default newline definition that was chosen when PCRE
585    was built. Setting the first or the second specifies that a newline is
586    indicated by a single character (CR or LF, respectively). Setting
587    PCRE_NEWLINE_CRLF specifies that a newline is indicated by the two-character
588    CRLF sequence. Setting PCRE_NEWLINE_ANYCRLF specifies that any of the three
589    preceding sequences should be recognized. Setting PCRE_NEWLINE_ANY specifies
590    that any Unicode newline sequence should be recognized. The Unicode newline
591    sequences are the three just mentioned, plus the single characters VT (vertical
592    tab, U+000B), FF (formfeed, U+000C), NEL (next line, U+0085), LS (line
593    separator, U+2028), and PS (paragraph separator, U+2029). The last two are
594    recognized only in UTF-8 mode.
595    </P>
596    <P>
597    The newline setting in the options word uses three bits that are treated
598    as a number, giving eight possibilities. Currently only six are used (default
599    plus the five values above). This means that if you set more than one newline
600    option, the combination may or may not be sensible. For example,
601    PCRE_NEWLINE_CR with PCRE_NEWLINE_LF is equivalent to PCRE_NEWLINE_CRLF, but
602    other combinations may yield unused numbers and cause an error.
603    </P>
604    <P>
605    The only time that a line break is specially recognized when compiling a
606    pattern is if PCRE_EXTENDED is set, and an unescaped # outside a character
607    class is encountered. This indicates a comment that lasts until after the next
608    line break sequence. In other circumstances, line break sequences are treated
609    as literal data, except that in PCRE_EXTENDED mode, both CR and LF are treated
610    as whitespace characters and are therefore ignored.
611  </P>  </P>
612  <P>  <P>
613    The newline option that is set at compile time becomes the default that is used
614    for <b>pcre_exec()</b> and <b>pcre_dfa_exec()</b>, but it can be overridden.
615  <pre>  <pre>
616    PCRE_NO_AUTO_CAPTURE    PCRE_NO_AUTO_CAPTURE
617  </PRE>  </pre>
 </P>  
 <P>  
618  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
619  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
620  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
621  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
622  in Perl.  in Perl.
 </P>  
 <P>  
623  <pre>  <pre>
624    PCRE_UNGREEDY    PCRE_UNGREEDY
625  </PRE>  </pre>
 </P>  
 <P>  
626  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
627  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
628  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>  
629  <pre>  <pre>
630    PCRE_UTF8    PCRE_UTF8
631  </PRE>  </pre>
 </P>  
 <P>  
632  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
633  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
634  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
635  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
636  behaviour of PCRE are given in the  behaviour of PCRE are given in the
637  <a href="pcre.html#utf8support">section on UTF-8 support</a>  <a href="pcre.html#utf8support">section on UTF-8 support</a>
638  in the main  in the main
639  <a href="pcre.html"><b>pcre</b></a>  <a href="pcre.html"><b>pcre</b></a>
640  page.  page.
641    <pre>
642      PCRE_NO_UTF8_CHECK
643    </pre>
644    When PCRE_UTF8 is set, the validity of the pattern as a UTF-8 string is
645    automatically checked. There is a discussion about the
646    <a href="pcre.html#utf8strings">validity of UTF-8 strings</a>
647    in the main
648    <a href="pcre.html"><b>pcre</b></a>
649    page. If an invalid UTF-8 sequence of bytes is found, <b>pcre_compile()</b>
650    returns an error. If you already know that your pattern is valid, and you want
651    to skip this check for performance reasons, you can set the PCRE_NO_UTF8_CHECK
652    option. When it is set, the effect of passing an invalid UTF-8 string as a
653    pattern is undefined. It may cause your program to crash. Note that this option
654    can also be passed to <b>pcre_exec()</b> and <b>pcre_dfa_exec()</b>, to suppress
655    the UTF-8 validity checking of subject strings.
656    </P>
657    <br><a name="SEC8" href="#TOC1">COMPILATION ERROR CODES</a><br>
658    <P>
659    The following table lists the error codes than may be returned by
660    <b>pcre_compile2()</b>, along with the error messages that may be returned by
661    both compiling functions. As PCRE has developed, some error codes have fallen
662    out of use. To avoid confusion, they have not been re-used.
663    <pre>
664       0  no error
665       1  \ at end of pattern
666       2  \c at end of pattern
667       3  unrecognized character follows \
668       4  numbers out of order in {} quantifier
669       5  number too big in {} quantifier
670       6  missing terminating ] for character class
671       7  invalid escape sequence in character class
672       8  range out of order in character class
673       9  nothing to repeat
674      10  [this code is not in use]
675      11  internal error: unexpected repeat
676      12  unrecognized character after (?
677      13  POSIX named classes are supported only within a class
678      14  missing )
679      15  reference to non-existent subpattern
680      16  erroffset passed as NULL
681      17  unknown option bit(s) set
682      18  missing ) after comment
683      19  [this code is not in use]
684      20  regular expression too large
685      21  failed to get memory
686      22  unmatched parentheses
687      23  internal error: code overflow
688      24  unrecognized character after (?&#60;
689      25  lookbehind assertion is not fixed length
690      26  malformed number or name after (?(
691      27  conditional group contains more than two branches
692      28  assertion expected after (?(
693      29  (?R or (?[+-]digits must be followed by )
694      30  unknown POSIX class name
695      31  POSIX collating elements are not supported
696      32  this version of PCRE is not compiled with PCRE_UTF8 support
697      33  [this code is not in use]
698      34  character value in \x{...} sequence is too large
699      35  invalid condition (?(0)
700      36  \C not allowed in lookbehind assertion
701      37  PCRE does not support \L, \l, \N, \U, or \u
702      38  number after (?C is &#62; 255
703      39  closing ) for (?C expected
704      40  recursive call could loop indefinitely
705      41  unrecognized character after (?P
706      42  syntax error in subpattern name (missing terminator)
707      43  two named subpatterns have the same name
708      44  invalid UTF-8 string
709      45  support for \P, \p, and \X has not been compiled
710      46  malformed \P or \p sequence
711      47  unknown property name after \P or \p
712      48  subpattern name is too long (maximum 32 characters)
713      49  too many named subpatterns (maximum 10,000)
714      50  [this code is not in use]
715      51  octal value is greater than \377 (not in UTF-8 mode)
716      52  internal error: overran compiling workspace
717      53  internal error: previously-checked referenced subpattern not found
718      54  DEFINE group contains more than one branch
719      55  repeating a DEFINE group is not allowed
720      56  inconsistent NEWLINE options
721      57  \g is not followed by a braced name or an optionally braced
722            non-zero number
723      58  (?+ or (?- or (?(+ or (?(- must be followed by a non-zero number
724    </PRE>
725  </P>  </P>
726  <br><a name="SEC6" href="#TOC1">STUDYING A PATTERN</a><br>  <br><a name="SEC9" href="#TOC1">STUDYING A PATTERN</a><br>
727  <P>  <P>
728  <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>
729  <b>const char **<i>errptr</i>);</b>  <b>const char **<i>errptr</i>);</b>
730  </P>  </P>
731  <P>  <P>
732  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
733  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
734  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
735  argument. If studing the pattern produces additional information that will help  argument. If studying the pattern produces additional information that will
736  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
737  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
738    results of the study.
739  </P>  </P>
740  <P>  <P>
741  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
742  <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
743  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
744  described below. If studying the pattern does not produce any additional  described
745  information, <b>pcre_study()</b> returns NULL. In that circumstance, if the  <a href="#extradata">below</a>
746  calling program wants to pass some of the other fields to <b>pcre_exec()</b>, it  in the section on matching a pattern.
 must set up its own <b>pcre_extra</b> block.  
747  </P>  </P>
748  <P>  <P>
749  The second argument contains option bits. At present, no options are defined  If studying the pattern does not produce any additional information
750  for <b>pcre_study()</b>, and this argument should always be zero.  <b>pcre_study()</b> returns NULL. In that circumstance, if the calling program
751    wants to pass any of the other fields to <b>pcre_exec()</b>, it must set up its
752    own <b>pcre_extra</b> block.
753    </P>
754    <P>
755    The second argument of <b>pcre_study()</b> contains option bits. At present, no
756    options are defined, and this argument should always be zero.
757  </P>  </P>
758  <P>  <P>
759  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
760  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
761  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
762  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
763  be sure that it has run successfully.  should test the error pointer for NULL after calling <b>pcre_study()</b>, to be
764    sure that it has run successfully.
765  </P>  </P>
766  <P>  <P>
767  This is a typical call to <b>pcre_study</b>():  This is a typical call to <b>pcre_study</b>():
 </P>  
 <P>  
768  <pre>  <pre>
769    pcre_extra *pe;    pcre_extra *pe;
770    pe = pcre_study(    pe = pcre_study(
771      re,             /* result of pcre_compile() */      re,             /* result of pcre_compile() */
772      0,              /* no options exist */      0,              /* no options exist */
773      &error);        /* set to NULL or points to a message */      &error);        /* set to NULL or points to a message */
774  </PRE>  </pre>
 </P>  
 <P>  
775  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
776  not have a single fixed starting character. A bitmap of possible starting  not have a single fixed starting character. A bitmap of possible starting
777  characters is created.  bytes is created.
778  </P>  <a name="localesupport"></a></P>
779  <a name="localesupport"></a><br><a name="SEC7" href="#TOC1">LOCALE SUPPORT</a><br>  <br><a name="SEC10" href="#TOC1">LOCALE SUPPORT</a><br>
780  <P>  <P>
781  PCRE handles caseless matching, and determines whether characters are letters,  PCRE handles caseless matching, and determines whether characters are letters,
782  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
783  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
784  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
785  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
786  is NULL, and is sufficient for many applications.  support. The use of locales with Unicode is discouraged. If you are handling
787  </P>  characters with codes greater than 128, you should either use UTF-8 and
788  <P>  Unicode, or use locales, but not try to mix the two.
789  An alternative set of tables can, however, be supplied. Such tables are built  </P>
790  by calling the <b>pcre_maketables()</b> function, which has no arguments, in the  <P>
791  relevant locale. The result can then be passed to <b>pcre_compile()</b> as often  PCRE contains an internal set of tables that are used when the final argument
792  as necessary. For example, to build and use tables that are appropriate for the  of <b>pcre_compile()</b> is NULL. These are sufficient for many applications.
793  French locale (where accented characters with codes greater than 128 are  Normally, the internal tables recognize only ASCII characters. However, when
794  treated as letters), the following code could be used:  PCRE is built, it is possible to cause the internal tables to be rebuilt in the
795  </P>  default "C" locale of the local system, which may cause them to be different.
796  <P>  </P>
797    <P>
798    The internal tables can always be overridden by tables supplied by the
799    application that calls PCRE. These may be created in a different locale from
800    the default. As more and more applications change to using Unicode, the need
801    for this locale support is expected to die away.
802    </P>
803    <P>
804    External tables are built by calling the <b>pcre_maketables()</b> function,
805    which has no arguments, in the relevant locale. The result can then be passed
806    to <b>pcre_compile()</b> or <b>pcre_exec()</b> as often as necessary. For
807    example, to build and use tables that are appropriate for the French locale
808    (where accented characters with values greater than 128 are treated as letters),
809    the following code could be used:
810  <pre>  <pre>
811    setlocale(LC_CTYPE, "fr");    setlocale(LC_CTYPE, "fr_FR");
812    tables = pcre_maketables();    tables = pcre_maketables();
813    re = pcre_compile(..., tables);    re = pcre_compile(..., tables);
814  </PRE>  </pre>
815    The locale name "fr_FR" is used on Linux and other Unix-like systems; if you
816    are using Windows, the name for the French locale is "french".
817  </P>  </P>
818  <P>  <P>
819  The tables are built in memory that is obtained via <b>pcre_malloc</b>. The  When <b>pcre_maketables()</b> runs, the tables are built in memory that is
820  pointer that is passed to <b>pcre_compile</b> is saved with the compiled  obtained via <b>pcre_malloc</b>. It is the caller's responsibility to ensure
821    that the memory containing the tables remains available for as long as it is
822    needed.
823    </P>
824    <P>
825    The pointer that is passed to <b>pcre_compile()</b> is saved with the compiled
826  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>
827  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
828  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
829  in different locales. It is the caller's responsibility to ensure that the  different patterns can be compiled in different locales.
830  memory containing the tables remains available for as long as it is needed.  </P>
831    <P>
832    It is possible to pass a table pointer or NULL (indicating the use of the
833    internal tables) to <b>pcre_exec()</b>. Although not intended for this purpose,
834    this facility could be used to match a pattern in a different locale from the
835    one in which it was compiled. Passing table pointers at run time is discussed
836    below in the section on matching a pattern.
837  </P>  </P>
838  <br><a name="SEC8" href="#TOC1">INFORMATION ABOUT A PATTERN</a><br>  <br><a name="SEC11" href="#TOC1">INFORMATION ABOUT A PATTERN</a><br>
839  <P>  <P>
840  <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>
841  <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 852  the pattern was not studied. The third a
852  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
853  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
854  the following negative numbers:  the following negative numbers:
 </P>  
 <P>  
855  <pre>  <pre>
856    PCRE_ERROR_NULL       the argument <i>code</i> was NULL    PCRE_ERROR_NULL       the argument <i>code</i> was NULL
857                          the argument <i>where</i> was NULL                          the argument <i>where</i> was NULL
858    PCRE_ERROR_BADMAGIC   the "magic number" was not found    PCRE_ERROR_BADMAGIC   the "magic number" was not found
859    PCRE_ERROR_BADOPTION  the value of <i>what</i> was invalid    PCRE_ERROR_BADOPTION  the value of <i>what</i> was invalid
860  </PRE>  </pre>
861  </P>  The "magic number" is placed at the start of each compiled pattern as an simple
862  <P>  check against passing an arbitrary memory pointer. Here is a typical call of
863  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>  
864  <pre>  <pre>
865    int rc;    int rc;
866    unsigned long int length;    size_t length;
867    rc = pcre_fullinfo(    rc = pcre_fullinfo(
868      re,               /* result of pcre_compile() */      re,               /* result of pcre_compile() */
869      pe,               /* result of pcre_study(), or NULL */      pe,               /* result of pcre_study(), or NULL */
870      PCRE_INFO_SIZE,   /* what is required */      PCRE_INFO_SIZE,   /* what is required */
871      &length);         /* where to put the data */      &length);         /* where to put the data */
872  </PRE>  </pre>
 </P>  
 <P>  
873  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
874  as follows:  as follows:
 </P>  
 <P>  
875  <pre>  <pre>
876    PCRE_INFO_BACKREFMAX    PCRE_INFO_BACKREFMAX
877  </PRE>  </pre>
 </P>  
 <P>  
878  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
879  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
880  no back references.  no back references.
 </P>  
 <P>  
881  <pre>  <pre>
882    PCRE_INFO_CAPTURECOUNT    PCRE_INFO_CAPTURECOUNT
883  </PRE>  </pre>
 </P>  
 <P>  
884  Return the number of capturing subpatterns in the pattern. The fourth argument  Return the number of capturing subpatterns in the pattern. The fourth argument
885  should point to an \fbint\fR variable.  should point to an <b>int</b> variable.
886  </P>  <pre>
887  <P>    PCRE_INFO_DEFAULT_TABLES
888    </pre>
889    Return a pointer to the internal default character tables within PCRE. The
890    fourth argument should point to an <b>unsigned char *</b> variable. This
891    information call is provided for internal use by the <b>pcre_study()</b>
892    function. External callers can cause PCRE to use its internal tables by passing
893    a NULL table pointer.
894  <pre>  <pre>
895    PCRE_INFO_FIRSTBYTE    PCRE_INFO_FIRSTBYTE
896  </PRE>  </pre>
 </P>  
 <P>  
897  Return information about the first byte of any matched string, for a  Return information about the first byte of any matched string, for a
898  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>
899  old name is still recognized for backwards compatibility.)  variable. (This option used to be called PCRE_INFO_FIRSTCHAR; the old name is
900  </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  
901  </P>  </P>
902  <P>  <P>
903    If there is a fixed first byte, for example, from a pattern such as
904    (cat|cow|coyote), its value is returned. Otherwise, if either
905    <br>
906    <br>
907  (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
908  starts with "^", or  starts with "^", or
909  </P>  <br>
910  <P>  <br>
911  (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
912  (if it were set, the pattern would be anchored),  (if it were set, the pattern would be anchored),
913  </P>  <br>
914  <P>  <br>
915  -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
916  subject string or after any newline within the string. Otherwise -2 is  subject string or after any newline within the string. Otherwise -2 is
917  returned. For anchored patterns, -2 is returned.  returned. For anchored patterns, -2 is returned.
 </P>  
 <P>  
918  <pre>  <pre>
919    PCRE_INFO_FIRSTTABLE    PCRE_INFO_FIRSTTABLE
920  </PRE>  </pre>
 </P>  
 <P>  
921  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
922  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
923  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
924  fourth argument should point to an <b>unsigned char *</b> variable.  fourth argument should point to an <b>unsigned char *</b> variable.
925  </P>  <pre>
926  <P>    PCRE_INFO_HASCRORLF
927    </pre>
928    Return 1 if the pattern contains any explicit matches for CR or LF characters,
929    otherwise 0. The fourth argument should point to an <b>int</b> variable. An
930    explicit match is either a literal CR or LF character, or \r or \n.
931    <pre>
932      PCRE_INFO_JCHANGED
933    </pre>
934    Return 1 if the (?J) option setting is used in the pattern, otherwise 0. The
935    fourth argument should point to an <b>int</b> variable. The (?J) internal option
936    setting changes the local PCRE_DUPNAMES option.
937  <pre>  <pre>
938    PCRE_INFO_LASTLITERAL    PCRE_INFO_LASTLITERAL
939  </PRE>  </pre>
 </P>  
 <P>  
940  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
941  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
942  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 944  returned. For anchored patterns, a last
944  follows something of variable length. For example, for the pattern  follows something of variable length. For example, for the pattern
945  /^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
946  is -1.  is -1.
 </P>  
 <P>  
947  <pre>  <pre>
948    PCRE_INFO_NAMECOUNT    PCRE_INFO_NAMECOUNT
949    PCRE_INFO_NAMEENTRYSIZE    PCRE_INFO_NAMEENTRYSIZE
950    PCRE_INFO_NAMETABLE    PCRE_INFO_NAMETABLE
951  </PRE>  </pre>
 </P>  
 <P>  
952  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
953  names are just an additional way of identifying the parentheses, which still  names are just an additional way of identifying the parentheses, which still
954  acquire a number. A caller that wants to extract data from a named subpattern  acquire numbers. Several convenience functions such as
955  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
956  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
957  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
958  table for the pattern.  output vector (described with <b>pcre_exec()</b> below). To do the conversion,
959    you need to use the name-to-number map, which is described by these three
960    values.
961  </P>  </P>
962  <P>  <P>
963  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 967  length of the longest name. PCRE_INFO_NA
967  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
968  are the number of the capturing parenthesis, most significant byte first. The  are the number of the capturing parenthesis, most significant byte first. The
969  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
970  alphabetical order. For example, consider the following pattern (assume  alphabetical order. When PCRE_DUPNAMES is set, duplicate names are in order of
971    their parentheses numbers. For example, consider the following pattern (assume
972  PCRE_EXTENDED is set, so white space - including newlines - is ignored):  PCRE_EXTENDED is set, so white space - including newlines - is ignored):
 </P>  
 <P>  
973  <pre>  <pre>
974    (?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) )
975    (?P&#60;month&#62;\d\d) - (?P&#60;day&#62;\d\d) )  </pre>
 </PRE>  
 </P>  
 <P>  
976  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
977  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
978  bytes shows in hex, and undefined bytes shown as ??:  bytes shows in hexadecimal, and undefined bytes shown as ??:
 </P>  
 <P>  
979  <pre>  <pre>
980    00 01 d  a  t  e  00 ??    00 01 d  a  t  e  00 ??
981    00 05 d  a  y  00 ?? ??    00 05 d  a  y  00 ?? ??
982    00 04 m  o  n  t  h  00    00 04 m  o  n  t  h  00
983    00 02 y  e  a  r  00 ??    00 02 y  e  a  r  00 ??
984  </PRE>  </pre>
985  </P>  When writing code to extract data from named subpatterns using the
986  <P>  name-to-number map, remember that the length of the entries is likely to be
987  When writing code to extract data from named subpatterns, remember that the  different for each compiled pattern.
988  length of each entry may be different for each compiled pattern.  <pre>
989  </P>    PCRE_INFO_OKPARTIAL
990  <P>  </pre>
991    Return 1 if the pattern can be used for partial matching, otherwise 0. The
992    fourth argument should point to an <b>int</b> variable. The
993    <a href="pcrepartial.html"><b>pcrepartial</b></a>
994    documentation lists the restrictions that apply to patterns when partial
995    matching is used.
996  <pre>  <pre>
997    PCRE_INFO_OPTIONS    PCRE_INFO_OPTIONS
998  </PRE>  </pre>
 </P>  
 <P>  
999  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
1000  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
1001  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
1002  top-level option settings within the pattern itself.  top-level option settings at the start of the pattern itself. In other words,
1003    they are the options that will be in force when matching starts. For example,
1004    if the pattern /(?im)abc(?-i)d/ is compiled with the PCRE_EXTENDED option, the
1005    result is PCRE_CASELESS, PCRE_MULTILINE, and PCRE_EXTENDED.
1006  </P>  </P>
1007  <P>  <P>
1008  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
1009  alternatives begin with one of the following:  alternatives begin with one of the following:
 </P>  
 <P>  
1010  <pre>  <pre>
1011    ^     unless PCRE_MULTILINE is set    ^     unless PCRE_MULTILINE is set
1012    \A    always    \A    always
1013    \G    always    \G    always
1014    .*    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
1015            references to the subpattern in which .* appears  </pre>
 </PRE>  
 </P>  
 <P>  
1016  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
1017  <b>pcre_fullinfo()</b>.  <b>pcre_fullinfo()</b>.
 </P>  
 <P>  
1018  <pre>  <pre>
1019    PCRE_INFO_SIZE    PCRE_INFO_SIZE
1020  </PRE>  </pre>
 </P>  
 <P>  
1021  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
1022  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
1023  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>
1024  variable.  variable.
 </P>  
 <P>  
1025  <pre>  <pre>
1026    PCRE_INFO_STUDYSIZE    PCRE_INFO_STUDYSIZE
1027  </PRE>  </pre>
1028  </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  
1029  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
1030  <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
1031  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
1032  <b>size_t</b> variable.  <b>size_t</b> variable.
1033  </P>  </P>
1034  <br><a name="SEC9" href="#TOC1">OBSOLETE INFO FUNCTION</a><br>  <br><a name="SEC12" href="#TOC1">OBSOLETE INFO FUNCTION</a><br>
1035  <P>  <P>
1036  <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>
1037  <b>*<i>firstcharptr</i>);</b>  <b>*<i>firstcharptr</i>);</b>
# Line 753  restrictive to return all the available Line 1042  restrictive to return all the available
1042  programs should use <b>pcre_fullinfo()</b> instead. The yield of  programs should use <b>pcre_fullinfo()</b> instead. The yield of
1043  <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
1044  following negative numbers:  following negative numbers:
 </P>  
 <P>  
1045  <pre>  <pre>
1046    PCRE_ERROR_NULL       the argument <i>code</i> was NULL    PCRE_ERROR_NULL       the argument <i>code</i> was NULL
1047    PCRE_ERROR_BADMAGIC   the "magic number" was not found    PCRE_ERROR_BADMAGIC   the "magic number" was not found
1048  </PRE>  </pre>
 </P>  
 <P>  
1049  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
1050  pattern was compiled is placed in the integer it points to (see  pattern was compiled is placed in the integer it points to (see
1051  PCRE_INFO_OPTIONS above).  PCRE_INFO_OPTIONS above).
# Line 770  If the pattern is not anchored and the < Line 1055  If the pattern is not anchored and the <
1055  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
1056  string (see PCRE_INFO_FIRSTBYTE above).  string (see PCRE_INFO_FIRSTBYTE above).
1057  </P>  </P>
1058  <br><a name="SEC10" href="#TOC1">MATCHING A PATTERN</a><br>  <br><a name="SEC13" href="#TOC1">REFERENCE COUNTS</a><br>
1059    <P>
1060    <b>int pcre_refcount(pcre *<i>code</i>, int <i>adjust</i>);</b>
1061    </P>
1062    <P>
1063    The <b>pcre_refcount()</b> function is used to maintain a reference count in the
1064    data block that contains a compiled pattern. It is provided for the benefit of
1065    applications that operate in an object-oriented manner, where different parts
1066    of the application may be using the same compiled pattern, but you want to free
1067    the block when they are all done.
1068    </P>
1069    <P>
1070    When a pattern is compiled, the reference count field is initialized to zero.
1071    It is changed only by calling this function, whose action is to add the
1072    <i>adjust</i> value (which may be positive or negative) to it. The yield of the
1073    function is the new value. However, the value of the count is constrained to
1074    lie between 0 and 65535, inclusive. If the new value is outside these limits,
1075    it is forced to the appropriate limit value.
1076    </P>
1077    <P>
1078    Except when it is zero, the reference count is not correctly preserved if a
1079    pattern is compiled on one host and then transferred to a host whose byte-order
1080    is different. (This seems a highly unlikely scenario.)
1081    </P>
1082    <br><a name="SEC14" href="#TOC1">MATCHING A PATTERN: THE TRADITIONAL FUNCTION</a><br>
1083  <P>  <P>
1084  <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>
1085  <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 1087  string (see PCRE_INFO_FIRSTBYTE above).
1087  </P>  </P>
1088  <P>  <P>
1089  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
1090  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
1091  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
1092  <i>extra</i> argument.  <i>extra</i> argument. This function is the main matching facility of the
1093    library, and it operates in a Perl-like manner. For specialist use there is
1094    also an alternative matching function, which is described
1095    <a href="#dfamatch">below</a>
1096    in the section about the <b>pcre_dfa_exec()</b> function.
1097  </P>  </P>
1098  <P>  <P>
1099  Here is an example of a simple call to <b>pcre_exec()</b>:  In most applications, the pattern will have been compiled (and optionally
1100    studied) in the same process that calls <b>pcre_exec()</b>. However, it is
1101    possible to save compiled patterns and study data, and then use them later
1102    in different processes, possibly even on different hosts. For a discussion
1103    about this, see the
1104    <a href="pcreprecompile.html"><b>pcreprecompile</b></a>
1105    documentation.
1106  </P>  </P>
1107  <P>  <P>
1108    Here is an example of a simple call to <b>pcre_exec()</b>:
1109  <pre>  <pre>
1110    int rc;    int rc;
1111    int ovector[30];    int ovector[30];
# Line 796  Here is an example of a simple call to < Line 1116  Here is an example of a simple call to <
1116      11,             /* the length of the subject string */      11,             /* the length of the subject string */
1117      0,              /* start at offset 0 in the subject */      0,              /* start at offset 0 in the subject */
1118      0,              /* default options */      0,              /* default options */
1119      ovector,        /* vector for substring information */      ovector,        /* vector of integers for substring information */
1120      30);            /* number of elements in the vector */      30);            /* number of elements (NOT size in bytes) */
1121  </PRE>  <a name="extradata"></a></PRE>
1122  </P>  </P>
1123    <br><b>
1124    Extra data for <b>pcre_exec()</b>
1125    </b><br>
1126  <P>  <P>
1127  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>
1128  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
1129  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
1130  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
1131  </P>  fields (not necessarily in this order):
 <P>  
1132  <pre>  <pre>
1133    unsigned long int <i>flags</i>;    unsigned long int <i>flags</i>;
1134    void *<i>study_data</i>;    void *<i>study_data</i>;
1135    unsigned long int <i>match_limit</i>;    unsigned long int <i>match_limit</i>;
1136      unsigned long int <i>match_limit_recursion</i>;
1137    void *<i>callout_data</i>;    void *<i>callout_data</i>;
1138  </PRE>    const unsigned char *<i>tables</i>;
1139  </P>  </pre>
 <P>  
1140  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
1141  are set. The flag bits are:  are set. The flag bits are:
 </P>  
 <P>  
1142  <pre>  <pre>
1143    PCRE_EXTRA_STUDY_DATA    PCRE_EXTRA_STUDY_DATA
1144    PCRE_EXTRA_MATCH_LIMIT    PCRE_EXTRA_MATCH_LIMIT
1145      PCRE_EXTRA_MATCH_LIMIT_RECURSION
1146    PCRE_EXTRA_CALLOUT_DATA    PCRE_EXTRA_CALLOUT_DATA
1147  </PRE>    PCRE_EXTRA_TABLES
1148  </P>  </pre>
 <P>  
1149  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
1150  <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
1151  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
1152  the block by setting the other fields.  the block by setting the other fields and their corresponding flag bits.
1153  </P>  </P>
1154  <P>  <P>
1155  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
1156  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,
1157  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
1158  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.  
1159  </P>  </P>
1160  <P>  <P>
1161  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
1162    (sometimes recursively). The limit set by <i>match_limit</i> is imposed on the
1163    number of times this function is called during a match, which has the effect of
1164    limiting the amount of backtracking that can take place. For patterns that are
1165    not anchored, the count restarts from zero for each position in the subject
1166    string.
1167    </P>
1168    <P>
1169    The default value for the limit can be set when PCRE is built; the default
1170  default is 10 million, which handles all but the most extreme cases. You can  default is 10 million, which handles all but the most extreme cases. You can
1171  reduce the default by suppling <b>pcre_exec()</b> with a \fRpcre_extra\fR block  override the default by suppling <b>pcre_exec()</b> with a <b>pcre_extra</b>
1172  in which <i>match_limit</i> is set to a smaller value, and  block in which <i>match_limit</i> is set, and PCRE_EXTRA_MATCH_LIMIT is set in
1173  PCRE_EXTRA_MATCH_LIMIT is set in the <i>flags</i> field. If the limit is  the <i>flags</i> field. If the limit is exceeded, <b>pcre_exec()</b> returns
1174  exceeded, <b>pcre_exec()</b> returns PCRE_ERROR_MATCHLIMIT.  PCRE_ERROR_MATCHLIMIT.
1175    </P>
1176    <P>
1177    The <i>match_limit_recursion</i> field is similar to <i>match_limit</i>, but
1178    instead of limiting the total number of times that <b>match()</b> is called, it
1179    limits the depth of recursion. The recursion depth is a smaller number than the
1180    total number of calls, because not all calls to <b>match()</b> are recursive.
1181    This limit is of use only if it is set smaller than <i>match_limit</i>.
1182    </P>
1183    <P>
1184    Limiting the recursion depth limits the amount of stack that can be used, or,
1185    when PCRE has been compiled to use memory on the heap instead of the stack, the
1186    amount of heap memory that can be used.
1187    </P>
1188    <P>
1189    The default value for <i>match_limit_recursion</i> can be set when PCRE is
1190    built; the default default is the same value as the default for
1191    <i>match_limit</i>. You can override the default by suppling <b>pcre_exec()</b>
1192    with a <b>pcre_extra</b> block in which <i>match_limit_recursion</i> is set, and
1193    PCRE_EXTRA_MATCH_LIMIT_RECURSION is set in the <i>flags</i> field. If the limit
1194    is exceeded, <b>pcre_exec()</b> returns PCRE_ERROR_RECURSIONLIMIT.
1195  </P>  </P>
1196  <P>  <P>
1197  The <i>pcre_callout</i> field is used in conjunction with the "callout" feature,  The <i>pcre_callout</i> field is used in conjunction with the "callout" feature,
1198  which is described in the <b>pcrecallout</b> documentation.  which is described in the
1199    <a href="pcrecallout.html"><b>pcrecallout</b></a>
1200    documentation.
1201  </P>  </P>
1202  <P>  <P>
1203  The PCRE_ANCHORED option can be passed in the <i>options</i> argument, whose  The <i>tables</i> field is used to pass a character tables pointer to
1204  unused bits must be zero. This limits <b>pcre_exec()</b> to matching at the  <b>pcre_exec()</b>; this overrides the value that is stored with the compiled
1205  first matching position. However, if a pattern was compiled with PCRE_ANCHORED,  pattern. A non-NULL value is stored with the compiled pattern only if custom
1206  or turned out to be anchored by virtue of its contents, it cannot be made  tables were supplied to <b>pcre_compile()</b> via its <i>tableptr</i> argument.
1207  unachored at matching time.  If NULL is passed to <b>pcre_exec()</b> using this mechanism, it forces PCRE's
1208    internal tables to be used. This facility is helpful when re-using patterns
1209    that have been saved after compiling with an external set of tables, because
1210    the external tables might be at a different address when <b>pcre_exec()</b> is
1211    called. See the
1212    <a href="pcreprecompile.html"><b>pcreprecompile</b></a>
1213    documentation for a discussion of saving compiled patterns for later use.
1214    <a name="execoptions"></a></P>
1215    <br><b>
1216    Option bits for <b>pcre_exec()</b>
1217    </b><br>
1218    <P>
1219    The unused bits of the <i>options</i> argument for <b>pcre_exec()</b> must be
1220    zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_<i>xxx</i>,
1221    PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NO_UTF8_CHECK and PCRE_PARTIAL.
1222    <pre>
1223      PCRE_ANCHORED
1224    </pre>
1225    The PCRE_ANCHORED option limits <b>pcre_exec()</b> to matching at the first
1226    matching position. If a pattern was compiled with PCRE_ANCHORED, or turned out
1227    to be anchored by virtue of its contents, it cannot be made unachored at
1228    matching time.
1229    <pre>
1230      PCRE_BSR_ANYCRLF
1231      PCRE_BSR_UNICODE
1232    </pre>
1233    These options (which are mutually exclusive) control what the \R escape
1234    sequence matches. The choice is either to match only CR, LF, or CRLF, or to
1235    match any Unicode newline sequence. These options override the choice that was
1236    made or defaulted when the pattern was compiled.
1237    <pre>
1238      PCRE_NEWLINE_CR
1239      PCRE_NEWLINE_LF
1240      PCRE_NEWLINE_CRLF
1241      PCRE_NEWLINE_ANYCRLF
1242      PCRE_NEWLINE_ANY
1243    </pre>
1244    These options override the newline definition that was chosen or defaulted when
1245    the pattern was compiled. For details, see the description of
1246    <b>pcre_compile()</b> above. During matching, the newline choice affects the
1247    behaviour of the dot, circumflex, and dollar metacharacters. It may also alter
1248    the way the match position is advanced after a match failure for an unanchored
1249    pattern.
1250  </P>  </P>
1251  <P>  <P>
1252  There are also three further options that can be set only at matching time:  When PCRE_NEWLINE_CRLF, PCRE_NEWLINE_ANYCRLF, or PCRE_NEWLINE_ANY is set, and a
1253    match attempt for an unanchored pattern fails when the current position is at a
1254    CRLF sequence, and the pattern contains no explicit matches for CR or LF
1255    characters, the match position is advanced by two characters instead of one, in
1256    other words, to after the CRLF.
1257  </P>  </P>
1258  <P>  <P>
1259  <pre>  The above rule is a compromise that makes the most common cases work as
1260    PCRE_NOTBOL  expected. For example, if the pattern is .+A (and the PCRE_DOTALL option is not
1261  </PRE>  set), it does not match the string "\r\nA" because, after failing at the
1262    start, it skips both the CR and the LF before retrying. However, the pattern
1263    [\r\n]A does match that string, because it contains an explicit CR or LF
1264    reference, and so advances only by one character after the first failure.
1265  </P>  </P>
1266  <P>  <P>
1267  The first character of the string is not the beginning of a line, so the  An explicit match for CR of LF is either a literal appearance of one of those
1268  circumflex metacharacter should not match before it. Setting this without  characters, or one of the \r or \n escape sequences. Implicit matches such as
1269  PCRE_MULTILINE (at compile time) causes circumflex never to match.  [^X] do not count, nor does \s (which includes CR and LF in the characters
1270    that it matches).
1271  </P>  </P>
1272  <P>  <P>
1273    Notwithstanding the above, anomalous effects may still occur when CRLF is a
1274    valid newline sequence and explicit \r or \n escapes appear in the pattern.
1275    <pre>
1276      PCRE_NOTBOL
1277    </pre>
1278    This option specifies that first character of the subject string is not the
1279    beginning of a line, so the circumflex metacharacter should not match before
1280    it. Setting this without PCRE_MULTILINE (at compile time) causes circumflex
1281    never to match. This option affects only the behaviour of the circumflex
1282    metacharacter. It does not affect \A.
1283  <pre>  <pre>
1284    PCRE_NOTEOL    PCRE_NOTEOL
1285  </PRE>  </pre>
1286  </P>  This option specifies that the end of the subject string is not the end of a
1287  <P>  line, so the dollar metacharacter should not match it nor (except in multiline
1288  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
1289  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
1290  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>  
1291  <pre>  <pre>
1292    PCRE_NOTEMPTY    PCRE_NOTEMPTY
1293  </PRE>  </pre>
 </P>  
 <P>  
1294  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
1295  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
1296  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>  
1297  <pre>  <pre>
1298    a?b?    a?b?
1299  </PRE>  </pre>
 </P>  
 <P>  
1300  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
1301  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
1302  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 1306  Perl has no direct equivalent of PCRE_NO
1306  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
1307  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
1308  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
1309  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
1310  below) and trying an ordinary match again.  starting offset (see below) and trying an ordinary match again. There is some
1311    code that demonstrates how to do this in the <i>pcredemo.c</i> sample program.
1312    <pre>
1313      PCRE_NO_UTF8_CHECK
1314    </pre>
1315    When PCRE_UTF8 is set at compile time, the validity of the subject as a UTF-8
1316    string is automatically checked when <b>pcre_exec()</b> is subsequently called.
1317    The value of <i>startoffset</i> is also checked to ensure that it points to the
1318    start of a UTF-8 character. There is a discussion about the validity of UTF-8
1319    strings in the
1320    <a href="pcre.html#utf8strings">section on UTF-8 support</a>
1321    in the main
1322    <a href="pcre.html"><b>pcre</b></a>
1323    page. If an invalid UTF-8 sequence of bytes is found, <b>pcre_exec()</b> returns
1324    the error PCRE_ERROR_BADUTF8. If <i>startoffset</i> contains an invalid value,
1325    PCRE_ERROR_BADUTF8_OFFSET is returned.
1326    </P>
1327    <P>
1328    If you already know that your subject is valid, and you want to skip these
1329    checks for performance reasons, you can set the PCRE_NO_UTF8_CHECK option when
1330    calling <b>pcre_exec()</b>. You might want to do this for the second and
1331    subsequent calls to <b>pcre_exec()</b> if you are making repeated calls to find
1332    all the matches in a single subject string. However, you should be sure that
1333    the value of <i>startoffset</i> points to the start of a UTF-8 character. When
1334    PCRE_NO_UTF8_CHECK is set, the effect of passing an invalid UTF-8 string as a
1335    subject, or a value of <i>startoffset</i> that does not point to the start of a
1336    UTF-8 character, is undefined. Your program may crash.
1337    <pre>
1338      PCRE_PARTIAL
1339    </pre>
1340    This option turns on the partial matching feature. If the subject string fails
1341    to match the pattern, but at some point during the matching process the end of
1342    the subject was reached (that is, the subject partially matches the pattern and
1343    the failure to match occurred only because there were not enough subject
1344    characters), <b>pcre_exec()</b> returns PCRE_ERROR_PARTIAL instead of
1345    PCRE_ERROR_NOMATCH. When PCRE_PARTIAL is used, there are restrictions on what
1346    may appear in the pattern. These are discussed in the
1347    <a href="pcrepartial.html"><b>pcrepartial</b></a>
1348    documentation.
1349  </P>  </P>
1350    <br><b>
1351    The string to be matched by <b>pcre_exec()</b>
1352    </b><br>
1353  <P>  <P>
1354  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
1355  <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
1356  <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
1357  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
1358  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
1359  </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.  
1360  </P>  </P>
1361  <P>  <P>
1362  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 1364  same subject by calling <b>pcre_exec()</
1364  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
1365  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
1366  lookbehind. For example, consider the pattern  lookbehind. For example, consider the pattern
 </P>  
 <P>  
1367  <pre>  <pre>
1368    \Biss\B    \Biss\B
1369  </PRE>  </pre>
 </P>  
 <P>  
1370  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
1371  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
1372  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 1379  behind the starting point to discover th
1379  </P>  </P>
1380  <P>  <P>
1381  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
1382  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
1383  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.
1384  </P>  </P>
1385    <br><b>
1386    How <b>pcre_exec()</b> returns captured substrings
1387    </b><br>
1388  <P>  <P>
1389  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
1390  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 1396  kinds of parenthesized subpattern that d
1396  <P>  <P>
1397  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
1398  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
1399  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>:
1400  back captured substrings, each substring using a pair of integers. The  this argument is NOT the size of <i>ovector</i> in bytes.
1401  remaining third of the vector is used as workspace by <b>pcre_exec()</b> while  </P>
1402  matching capturing subpatterns, and is not available for passing back  <P>
1403  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,
1404  three. If it is not, it is rounded down.  each substring using a pair of integers. The remaining third of the vector is
1405    used as workspace by <b>pcre_exec()</b> while matching capturing subpatterns,
1406    and is not available for passing back information. The length passed in
1407    <i>ovecsize</i> should always be a multiple of three. If it is not, it is
1408    rounded down.
1409  </P>  </P>
1410  <P>  <P>
1411  When a match has been successful, information about captured substrings is  When a match is successful, information about captured substrings is returned
1412  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
1413  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
1414  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
1415  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
1416  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
1417  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
1418  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>
1419  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
1420  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
1421  just the first pair of offsets has been set.  capturing subpatterns, the return value from a successful match is 1,
1422  </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.  
1423  </P>  </P>
1424  <P>  <P>
1425  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
1426  string that it matched that gets returned.  string that it matched that is returned.
1427  </P>  </P>
1428  <P>  <P>
1429  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
1430  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
1431  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
1432  <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
1433  <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
1434  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
1435  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
1436  to supply an <i>ovector</i>.  advisable to supply an <i>ovector</i>.
1437  </P>  </P>
1438  <P>  <P>
1439  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
1440  subpatterns there are in a compiled pattern. The smallest size for  subpatterns there are in a compiled pattern. The smallest size for
1441  <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
1442  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.
1443  </P>  </P>
1444  <P>  <P>
1445  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
1446  defined in the header file:  the subject when subpattern <i>n</i> has not been used at all. For example, if
1447    the string "abc" is matched against the pattern (a|(z))(bc) the return from the
1448    function is 4, and subpatterns 1 and 3 are matched, but 2 is not. When this
1449    happens, both values in the offset pairs corresponding to unused subpatterns
1450    are set to -1.
1451  </P>  </P>
1452  <P>  <P>
1453  <pre>  Offset values that correspond to unused subpatterns at the end of the
1454    PCRE_ERROR_NOMATCH        (-1)  expression are also set to -1. For example, if the string "abc" is matched
1455  </PRE>  against the pattern (abc)(x(yz)?)? subpatterns 2 and 3 are not matched. The
1456    return from the function is 2, because the highest used capturing subpattern
1457    number is 1. However, you can refer to the offsets for the second and third
1458    capturing subpatterns if you wish (assuming the vector is large enough, of
1459    course).
1460  </P>  </P>
1461  <P>  <P>
1462  The subject string did not match the pattern.  Some convenience functions are provided for extracting the captured substrings
1463  </P>  as separate strings. These are described below.
1464    <a name="errorlist"></a></P>
1465    <br><b>
1466    Error return values from <b>pcre_exec()</b>
1467    </b><br>
1468  <P>  <P>
1469    If <b>pcre_exec()</b> fails, it returns a negative number. The following are
1470    defined in the header file:
1471    <pre>
1472      PCRE_ERROR_NOMATCH        (-1)
1473    </pre>
1474    The subject string did not match the pattern.
1475  <pre>  <pre>
1476    PCRE_ERROR_NULL           (-2)    PCRE_ERROR_NULL           (-2)
1477  </PRE>  </pre>
 </P>  
 <P>  
1478  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
1479  NULL and <i>ovecsize</i> was not zero.  NULL and <i>ovecsize</i> was not zero.
 </P>  
 <P>  
1480  <pre>  <pre>
1481    PCRE_ERROR_BADOPTION      (-3)    PCRE_ERROR_BADOPTION      (-3)
1482  </PRE>  </pre>
 </P>  
 <P>  
1483  An unrecognized bit was set in the <i>options</i> argument.  An unrecognized bit was set in the <i>options</i> argument.
 </P>  
 <P>  
1484  <pre>  <pre>
1485    PCRE_ERROR_BADMAGIC       (-4)    PCRE_ERROR_BADMAGIC       (-4)
1486  </PRE>  </pre>
 </P>  
 <P>  
1487  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
1488  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
1489  magic number isn't present.  compiled in an environment of one endianness is run in an environment with the
1490  </P>  other endianness. This is the error that PCRE gives when the magic number is
1491  <P>  not present.
1492  <pre>  <pre>
1493    PCRE_ERROR_UNKNOWN_NODE   (-5)    PCRE_ERROR_UNKNOWN_OPCODE (-5)
1494  </PRE>  </pre>
 </P>  
 <P>  
1495  While running the pattern match, an unknown item was encountered in the  While running the pattern match, an unknown item was encountered in the
1496  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
1497  of the compiled pattern.  of the compiled pattern.
 </P>  
 <P>  
1498  <pre>  <pre>
1499    PCRE_ERROR_NOMEMORY       (-6)    PCRE_ERROR_NOMEMORY       (-6)
1500  </PRE>  </pre>
 </P>  
 <P>  
1501  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
1502  <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
1503  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
1504  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
1505  the end of matching.  automatically freed at the end of matching.
 </P>  
 <P>  
1506  <pre>  <pre>
1507    PCRE_ERROR_NOSUBSTRING    (-7)    PCRE_ERROR_NOSUBSTRING    (-7)
1508  </PRE>  </pre>
 </P>  
 <P>  
1509  This error is used by the <b>pcre_copy_substring()</b>,  This error is used by the <b>pcre_copy_substring()</b>,
1510  <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
1511  below). It is never returned by <b>pcre_exec()</b>.  below). It is never returned by <b>pcre_exec()</b>.
 </P>  
 <P>  
1512  <pre>  <pre>
1513    PCRE_ERROR_MATCHLIMIT     (-8)    PCRE_ERROR_MATCHLIMIT     (-8)
1514  </PRE>  </pre>
1515  </P>  The backtracking limit, as specified by the <i>match_limit</i> field in a
1516  <P>  <b>pcre_extra</b> structure (or defaulted) was reached. See the description
1517  The recursion and backtracking limit, as specified by the <i>match_limit</i>  above.
1518    <pre>
1519      PCRE_ERROR_CALLOUT        (-9)
1520    </pre>
1521    This error is never generated by <b>pcre_exec()</b> itself. It is provided for
1522    use by callout functions that want to yield a distinctive error code. See the
1523    <a href="pcrecallout.html"><b>pcrecallout</b></a>
1524    documentation for details.
1525    <pre>
1526      PCRE_ERROR_BADUTF8        (-10)
1527    </pre>
1528    A string that contains an invalid UTF-8 byte sequence was passed as a subject.
1529    <pre>
1530      PCRE_ERROR_BADUTF8_OFFSET (-11)
1531    </pre>
1532    The UTF-8 byte sequence that was passed as a subject was valid, but the value
1533    of <i>startoffset</i> did not point to the beginning of a UTF-8 character.
1534    <pre>
1535      PCRE_ERROR_PARTIAL        (-12)
1536    </pre>
1537    The subject string did not match, but it did match partially. See the
1538    <a href="pcrepartial.html"><b>pcrepartial</b></a>
1539    documentation for details of partial matching.
1540    <pre>
1541      PCRE_ERROR_BADPARTIAL     (-13)
1542    </pre>
1543    The PCRE_PARTIAL option was used with a compiled pattern containing items that
1544    are not supported for partial matching. See the
1545    <a href="pcrepartial.html"><b>pcrepartial</b></a>
1546    documentation for details of partial matching.
1547    <pre>
1548      PCRE_ERROR_INTERNAL       (-14)
1549    </pre>
1550    An unexpected internal error has occurred. This error could be caused by a bug
1551    in PCRE or by overwriting of the compiled pattern.
1552    <pre>
1553      PCRE_ERROR_BADCOUNT       (-15)
1554    </pre>
1555    This error is given if the value of the <i>ovecsize</i> argument is negative.
1556    <pre>
1557      PCRE_ERROR_RECURSIONLIMIT (-21)
1558    </pre>
1559    The internal recursion limit, as specified by the <i>match_limit_recursion</i>
1560  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
1561  description above.  description above.
 </P>  
 <P>  
1562  <pre>  <pre>
1563    PCRE_ERROR_CALLOUT        (-9)    PCRE_ERROR_BADNEWLINE     (-23)
1564  </PRE>  </pre>
1565    An invalid combination of PCRE_NEWLINE_<i>xxx</i> options was given.
1566  </P>  </P>
1567  <P>  <P>
1568  This error is never generated by <b>pcre_exec()</b> itself. It is provided for  Error numbers -16 to -20 and -22 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.  
1569  </P>  </P>
1570  <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>
1571  <P>  <P>
1572  <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>
1573  <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 1589  Captured substrings can be accessed dire
1589  <b>pcre_get_substring_list()</b> are provided for extracting captured substrings  <b>pcre_get_substring_list()</b> are provided for extracting captured substrings
1590  as new, separate, zero-terminated strings. These functions identify substrings  as new, separate, zero-terminated strings. These functions identify substrings
1591  by number. The next section describes functions for extracting named  by number. The next section describes functions for extracting named
1592  substrings. A substring that contains a binary zero is correctly extracted and  substrings.
1593  has a further zero added on the end, but the result is not, of course,  </P>
1594  a C string.  <P>
1595    A substring that contains a binary zero is correctly extracted and has a
1596    further zero added on the end, but the result is not, of course, a C string.
1597    However, you can process such a string by referring to the length that is
1598    returned by <b>pcre_copy_substring()</b> and <b>pcre_get_substring()</b>.
1599    Unfortunately, the interface to <b>pcre_get_substring_list()</b> is not adequate
1600    for handling strings containing binary zeros, because the end of the final
1601    string is not independently indicated.
1602  </P>  </P>
1603  <P>  <P>
1604  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:
1605  <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,
1606  <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
1607  <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
1608  captured by the match, including the substring that matched the entire regular  captured by the match, including the substring that matched the entire regular
1609  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
1610  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
1611  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
1612  the vector divided by three.  number of elements in the vector divided by three.
1613  </P>  </P>
1614  <P>  <P>
1615  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>
1616  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
1617  value of zero extracts the substring that matched the entire pattern, while  value of zero extracts the substring that matched the entire pattern, whereas
1618  higher values extract the captured substrings. For <b>pcre_copy_substring()</b>,  higher values extract the captured substrings. For <b>pcre_copy_substring()</b>,
1619  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
1620  <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
1621  obtained via <b>pcre_malloc</b>, and its address is returned via  obtained via <b>pcre_malloc</b>, and its address is returned via
1622  <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
1623  including the terminating zero, or one of  including the terminating zero, or one of these error codes:
 </P>  
 <P>  
1624  <pre>  <pre>
1625    PCRE_ERROR_NOMEMORY       (-6)    PCRE_ERROR_NOMEMORY       (-6)
1626  </PRE>  </pre>
 </P>  
 <P>  
1627  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
1628  memory failed for <b>pcre_get_substring()</b>.  memory failed for <b>pcre_get_substring()</b>.
 </P>  
 <P>  
1629  <pre>  <pre>
1630    PCRE_ERROR_NOSUBSTRING    (-7)    PCRE_ERROR_NOSUBSTRING    (-7)
1631  </PRE>  </pre>
 </P>  
 <P>  
1632  There is no substring whose number is <i>stringnumber</i>.  There is no substring whose number is <i>stringnumber</i>.
1633  </P>  </P>
1634  <P>  <P>
1635  The <b>pcre_get_substring_list()</b> function extracts all available substrings  The <b>pcre_get_substring_list()</b> function extracts all available substrings
1636  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
1637  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
1638  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
1639  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
1640  function is zero if all went well, or  function is zero if all went well, or the error code
 </P>  
 <P>  
1641  <pre>  <pre>
1642    PCRE_ERROR_NOMEMORY       (-6)    PCRE_ERROR_NOMEMORY       (-6)
1643  </PRE>  </pre>
 </P>  
 <P>  
1644  if the attempt to get the memory block failed.  if the attempt to get the memory block failed.
1645  </P>  </P>
1646  <P>  <P>
# Line 1202  a previous call of <b>pcre_get_substring Line 1658  a previous call of <b>pcre_get_substring
1658  <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
1659  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
1660  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
1661  linked via a special interface to another programming language which cannot use  linked via a special interface to another programming language that cannot use
1662  <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
1663  provided.  provided.
1664  </P>  </P>
1665  <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>
1666    <P>
1667    <b>int pcre_get_stringnumber(const pcre *<i>code</i>,</b>
1668    <b>const char *<i>name</i>);</b>
1669    </P>
1670  <P>  <P>
1671  <b>int pcre_copy_named_substring(const pcre *<i>code</i>,</b>  <b>int pcre_copy_named_substring(const pcre *<i>code</i>,</b>
1672  <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 1674  provided.
1674  <b>char *<i>buffer</i>, int <i>buffersize</i>);</b>  <b>char *<i>buffer</i>, int <i>buffersize</i>);</b>
1675  </P>  </P>
1676  <P>  <P>
 <b>int pcre_get_stringnumber(const pcre *<i>code</i>,</b>  
 <b>const char *<i>name</i>);</b>  
 </P>  
 <P>  
1677  <b>int pcre_get_named_substring(const pcre *<i>code</i>,</b>  <b>int pcre_get_named_substring(const pcre *<i>code</i>,</b>
1678  <b>const char *<i>subject</i>, int *<i>ovector</i>,</b>  <b>const char *<i>subject</i>, int *<i>ovector</i>,</b>
1679  <b>int <i>stringcount</i>, const char *<i>stringname</i>,</b>  <b>int <i>stringcount</i>, const char *<i>stringname</i>,</b>
1680  <b>const char **<i>stringptr</i>);</b>  <b>const char **<i>stringptr</i>);</b>
1681  </P>  </P>
1682  <P>  <P>
1683  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.
1684  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>  
1685  <pre>  <pre>
1686    ab(?&#60;xxx&#62;\d+)...    (a+)b(?&#60;xxx&#62;\d+)...
1687  </PRE>  </pre>
1688  </P>  the number of the subpattern called "xxx" is 2. If the name is known to be
1689  <P>  unique (PCRE_DUPNAMES was not set), you can find the number from the name by
1690  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
1691  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
1692  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
1693  whole job.  that name.
1694  </P>  </P>
1695  <P>  <P>
1696  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
1697  <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
1698  extract by number, and so are not re-described here. There are just two  two functions that do the whole job.
1699  differences.  </P>
1700    <P>
1701    Most of the arguments of <b>pcre_copy_named_substring()</b> and
1702    <b>pcre_get_named_substring()</b> are the same as those for the similarly named
1703    functions that extract by number. As these are described in the previous
1704    section, they are not re-described here. There are just two differences:
1705  </P>  </P>
1706  <P>  <P>
1707  First, instead of a substring number, a substring name is given. Second, there  First, instead of a substring number, a substring name is given. Second, there
# Line 1253  translation table. Line 1711  translation table.
1711  </P>  </P>
1712  <P>  <P>
1713  These functions call <b>pcre_get_stringnumber()</b>, and if it succeeds, they  These functions call <b>pcre_get_stringnumber()</b>, and if it succeeds, they
1714  then call <i>pcre_copy_substring()</i> or <i>pcre_get_substring()</i>, as  then call <b>pcre_copy_substring()</b> or <b>pcre_get_substring()</b>, as
1715  appropriate.  appropriate. <b>NOTE:</b> If PCRE_DUPNAMES is set and there are duplicate names,
1716    the behaviour may not be what you want (see the next section).
1717    </P>
1718    <br><a name="SEC17" href="#TOC1">DUPLICATE SUBPATTERN NAMES</a><br>
1719    <P>
1720    <b>int pcre_get_stringtable_entries(const pcre *<i>code</i>,</b>
1721    <b>const char *<i>name</i>, char **<i>first</i>, char **<i>last</i>);</b>
1722    </P>
1723    <P>
1724    When a pattern is compiled with the PCRE_DUPNAMES option, names for subpatterns
1725    are not required to be unique. Normally, patterns with duplicate names are such
1726    that in any one match, only one of the named subpatterns participates. An
1727    example is shown in the
1728    <a href="pcrepattern.html"><b>pcrepattern</b></a>
1729    documentation.
1730  </P>  </P>
1731  <P>  <P>
1732  Last updated: 03 February 2003  When duplicates are present, <b>pcre_copy_named_substring()</b> and
1733    <b>pcre_get_named_substring()</b> return the first substring corresponding to
1734    the given name that is set. If none are set, PCRE_ERROR_NOSUBSTRING (-7) is
1735    returned; no data is returned. The <b>pcre_get_stringnumber()</b> function
1736    returns one of the numbers that are associated with the name, but it is not
1737    defined which it is.
1738    </P>
1739    <P>
1740    If you want to get full details of all captured substrings for a given name,
1741    you must use the <b>pcre_get_stringtable_entries()</b> function. The first
1742    argument is the compiled pattern, and the second is the name. The third and
1743    fourth are pointers to variables which are updated by the function. After it
1744    has run, they point to the first and last entries in the name-to-number table
1745    for the given name. The function itself returns the length of each entry, or
1746    PCRE_ERROR_NOSUBSTRING (-7) if there are none. The format of the table is
1747    described above in the section entitled <i>Information about a pattern</i>.
1748    Given all the relevant entries for the name, you can extract each of their
1749    numbers, and hence the captured data, if any.
1750    </P>
1751    <br><a name="SEC18" href="#TOC1">FINDING ALL POSSIBLE MATCHES</a><br>
1752    <P>
1753    The traditional matching function uses a similar algorithm to Perl, which stops
1754    when it finds the first match, starting at a given point in the subject. If you
1755    want to find all possible matches, or the longest possible match, consider
1756    using the alternative matching function (see below) instead. If you cannot use
1757    the alternative function, but still need to find all possible matches, you
1758    can kludge it up by making use of the callout facility, which is described in
1759    the
1760    <a href="pcrecallout.html"><b>pcrecallout</b></a>
1761    documentation.
1762    </P>
1763    <P>
1764    What you have to do is to insert a callout right at the end of the pattern.
1765    When your callout function is called, extract and save the current matched
1766    substring. Then return 1, which forces <b>pcre_exec()</b> to backtrack and try
1767    other alternatives. Ultimately, when it runs out of matches, <b>pcre_exec()</b>
1768    will yield PCRE_ERROR_NOMATCH.
1769    <a name="dfamatch"></a></P>
1770    <br><a name="SEC19" href="#TOC1">MATCHING A PATTERN: THE ALTERNATIVE FUNCTION</a><br>
1771    <P>
1772    <b>int pcre_dfa_exec(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>
1773    <b>const char *<i>subject</i>, int <i>length</i>, int <i>startoffset</i>,</b>
1774    <b>int <i>options</i>, int *<i>ovector</i>, int <i>ovecsize</i>,</b>
1775    <b>int *<i>workspace</i>, int <i>wscount</i>);</b>
1776    </P>
1777    <P>
1778    The function <b>pcre_dfa_exec()</b> is called to match a subject string against
1779    a compiled pattern, using a matching algorithm that scans the subject string
1780    just once, and does not backtrack. This has different characteristics to the
1781    normal algorithm, and is not compatible with Perl. Some of the features of PCRE
1782    patterns are not supported. Nevertheless, there are times when this kind of
1783    matching can be useful. For a discussion of the two matching algorithms, see
1784    the
1785    <a href="pcrematching.html"><b>pcrematching</b></a>
1786    documentation.
1787    </P>
1788    <P>
1789    The arguments for the <b>pcre_dfa_exec()</b> function are the same as for
1790    <b>pcre_exec()</b>, plus two extras. The <i>ovector</i> argument is used in a
1791    different way, and this is described below. The other common arguments are used
1792    in the same way as for <b>pcre_exec()</b>, so their description is not repeated
1793    here.
1794    </P>
1795    <P>
1796    The two additional arguments provide workspace for the function. The workspace
1797    vector should contain at least 20 elements. It is used for keeping track of
1798    multiple paths through the pattern tree. More workspace will be needed for
1799    patterns and subjects where there are a lot of potential matches.
1800    </P>
1801    <P>
1802    Here is an example of a simple call to <b>pcre_dfa_exec()</b>:
1803    <pre>
1804      int rc;
1805      int ovector[10];
1806      int wspace[20];
1807      rc = pcre_dfa_exec(
1808        re,             /* result of pcre_compile() */
1809        NULL,           /* we didn't study the pattern */
1810        "some string",  /* the subject string */
1811        11,             /* the length of the subject string */
1812        0,              /* start at offset 0 in the subject */
1813        0,              /* default options */
1814        ovector,        /* vector of integers for substring information */
1815        10,             /* number of elements (NOT size in bytes) */
1816        wspace,         /* working space vector */
1817        20);            /* number of elements (NOT size in bytes) */
1818    </PRE>
1819    </P>
1820    <br><b>
1821    Option bits for <b>pcre_dfa_exec()</b>
1822    </b><br>
1823    <P>
1824    The unused bits of the <i>options</i> argument for <b>pcre_dfa_exec()</b> must be
1825    zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_<i>xxx</i>,
1826    PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NO_UTF8_CHECK, PCRE_PARTIAL,
1827    PCRE_DFA_SHORTEST, and PCRE_DFA_RESTART. All but the last three of these are
1828    the same as for <b>pcre_exec()</b>, so their description is not repeated here.
1829    <pre>
1830      PCRE_PARTIAL
1831    </pre>
1832    This has the same general effect as it does for <b>pcre_exec()</b>, but the
1833    details are slightly different. When PCRE_PARTIAL is set for
1834    <b>pcre_dfa_exec()</b>, the return code PCRE_ERROR_NOMATCH is converted into
1835    PCRE_ERROR_PARTIAL if the end of the subject is reached, there have been no
1836    complete matches, but there is still at least one matching possibility. The
1837    portion of the string that provided the partial match is set as the first
1838    matching string.
1839    <pre>
1840      PCRE_DFA_SHORTEST
1841    </pre>
1842    Setting the PCRE_DFA_SHORTEST option causes the matching algorithm to stop as
1843    soon as it has found one match. Because of the way the alternative algorithm
1844    works, this is necessarily the shortest possible match at the first possible
1845    matching point in the subject string.
1846    <pre>
1847      PCRE_DFA_RESTART
1848    </pre>
1849    When <b>pcre_dfa_exec()</b> is called with the PCRE_PARTIAL option, and returns
1850    a partial match, it is possible to call it again, with additional subject
1851    characters, and have it continue with the same match. The PCRE_DFA_RESTART
1852    option requests this action; when it is set, the <i>workspace</i> and
1853    <i>wscount</i> options must reference the same vector as before because data
1854    about the match so far is left in them after a partial match. There is more
1855    discussion of this facility in the
1856    <a href="pcrepartial.html"><b>pcrepartial</b></a>
1857    documentation.
1858    </P>
1859    <br><b>
1860    Successful returns from <b>pcre_dfa_exec()</b>
1861    </b><br>
1862    <P>
1863    When <b>pcre_dfa_exec()</b> succeeds, it may have matched more than one
1864    substring in the subject. Note, however, that all the matches from one run of
1865    the function start at the same point in the subject. The shorter matches are
1866    all initial substrings of the longer matches. For example, if the pattern
1867    <pre>
1868      &#60;.*&#62;
1869    </pre>
1870    is matched against the string
1871    <pre>
1872      This is &#60;something&#62; &#60;something else&#62; &#60;something further&#62; no more
1873    </pre>
1874    the three matched strings are
1875    <pre>
1876      &#60;something&#62;
1877      &#60;something&#62; &#60;something else&#62;
1878      &#60;something&#62; &#60;something else&#62; &#60;something further&#62;
1879    </pre>
1880    On success, the yield of the function is a number greater than zero, which is
1881    the number of matched substrings. The substrings themselves are returned in
1882    <i>ovector</i>. Each string uses two elements; the first is the offset to the
1883    start, and the second is the offset to the end. In fact, all the strings have
1884    the same start offset. (Space could have been saved by giving this only once,
1885    but it was decided to retain some compatibility with the way <b>pcre_exec()</b>
1886    returns data, even though the meaning of the strings is different.)
1887    </P>
1888    <P>
1889    The strings are returned in reverse order of length; that is, the longest
1890    matching string is given first. If there were too many matches to fit into
1891    <i>ovector</i>, the yield of the function is zero, and the vector is filled with
1892    the longest matches.
1893    </P>
1894    <br><b>
1895    Error returns from <b>pcre_dfa_exec()</b>
1896    </b><br>
1897    <P>
1898    The <b>pcre_dfa_exec()</b> function returns a negative number when it fails.
1899    Many of the errors are the same as for <b>pcre_exec()</b>, and these are
1900    described
1901    <a href="#errorlist">above.</a>
1902    There are in addition the following errors that are specific to
1903    <b>pcre_dfa_exec()</b>:
1904    <pre>
1905      PCRE_ERROR_DFA_UITEM      (-16)
1906    </pre>
1907    This return is given if <b>pcre_dfa_exec()</b> encounters an item in the pattern
1908    that it does not support, for instance, the use of \C or a back reference.
1909    <pre>
1910      PCRE_ERROR_DFA_UCOND      (-17)
1911    </pre>
1912    This return is given if <b>pcre_dfa_exec()</b> encounters a condition item that
1913    uses a back reference for the condition, or a test for recursion in a specific
1914    group. These are not supported.
1915    <pre>
1916      PCRE_ERROR_DFA_UMLIMIT    (-18)
1917    </pre>
1918    This return is given if <b>pcre_dfa_exec()</b> is called with an <i>extra</i>
1919    block that contains a setting of the <i>match_limit</i> field. This is not
1920    supported (it is meaningless).
1921    <pre>
1922      PCRE_ERROR_DFA_WSSIZE     (-19)
1923    </pre>
1924    This return is given if <b>pcre_dfa_exec()</b> runs out of space in the
1925    <i>workspace</i> vector.
1926    <pre>
1927      PCRE_ERROR_DFA_RECURSE    (-20)
1928    </pre>
1929    When a recursive subpattern is processed, the matching function calls itself
1930    recursively, using private vectors for <i>ovector</i> and <i>workspace</i>. This
1931    error is given if the output vector is not large enough. This should be
1932    extremely rare, as a vector of size 1000 is used.
1933    </P>
1934    <br><a name="SEC20" href="#TOC1">SEE ALSO</a><br>
1935    <P>
1936    <b>pcrebuild</b>(3), <b>pcrecallout</b>(3), <b>pcrecpp(3)</b>(3),
1937    <b>pcrematching</b>(3), <b>pcrepartial</b>(3), <b>pcreposix</b>(3),
1938    <b>pcreprecompile</b>(3), <b>pcresample</b>(3), <b>pcrestack</b>(3).
1939    </P>
1940    <br><a name="SEC21" href="#TOC1">AUTHOR</a><br>
1941    <P>
1942    Philip Hazel
1943    <br>
1944    University Computing Service
1945    <br>
1946    Cambridge CB2 3QH, England.
1947    <br>
1948    </P>
1949    <br><a name="SEC22" href="#TOC1">REVISION</a><br>
1950    <P>
1951    Last updated: 11 September 2007
1952    <br>
1953    Copyright &copy; 1997-2007 University of Cambridge.
1954  <br>  <br>
1955  Copyright &copy; 1997-2003 University of Cambridge.  <p>
1956    Return to the <a href="index.html">PCRE index page</a>.
1957    </p>

Legend:
Removed from v.65  
changed lines
  Added in v.243

  ViewVC Help
Powered by ViewVC 1.1.5