/[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 73 by nigel, Sat Feb 24 21:40:30 2007 UTC revision 227 by ph10, Tue Aug 21 15:00:15 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 106  conversion went wrong.<br> Line 141  conversion went wrong.<br>
141  <P>  <P>
142  <b>int (*pcre_callout)(pcre_callout_block *);</b>  <b>int (*pcre_callout)(pcre_callout_block *);</b>
143  </P>  </P>
144  <br><a name="SEC2" href="#TOC1">PCRE API</a><br>  <br><a name="SEC2" href="#TOC1">PCRE API OVERVIEW</a><br>
 <P>  
 PCRE has its own native API, which is described in this document. There is also  
 a set of wrapper functions that correspond to the POSIX regular expression API.  
 These are described in the <b>pcreposix</b> documentation.  
 </P>  
 <P>  
 The native API function prototypes are defined in the header file <b>pcre.h</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.  
 </P>  
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.
# Line 165  should be done before calling any PCRE f Line 225  should be done before calling any PCRE f
225  The global variables <b>pcre_stack_malloc</b> and <b>pcre_stack_free</b> are also  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  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  only when PCRE is compiled to use the heap for remembering data, instead of
228  recursive function calls. This is a non-standard way of building PCRE, for use  recursive function calls, when running the <b>pcre_exec()</b> function. See the
229  in environments that have limited stacks. Because of the greater use of memory  <a href="pcrebuild.html"><b>pcrebuild</b></a>
230  management, it runs more slowly. Separate functions are provided so that  documentation for details of how to do this. It is a non-standard way of
231  special-purpose external code can be used for this case. When used, these  building PCRE, for use in environments that have limited stacks. Because of the
232  functions are always called in a stack-like manner (last obtained, first  greater use of memory management, it runs more slowly. Separate functions are
233  freed), and always for memory blocks of the same size.  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>  </P>
240  <P>  <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. The choice of newline convention does not affect the interpretation of
280    the \n or \r escape sequences.
281  </P>  </P>
282  <br><a name="SEC3" href="#TOC1">MULTITHREADING</a><br>  <br><a name="SEC4" href="#TOC1">MULTITHREADING</a><br>
283  <P>  <P>
284  The PCRE functions can be used in multi-threading applications, with the  The PCRE functions can be used in multi-threading applications, with the
285  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>,
# Line 189  callout function pointed to by <b>pcre_c Line 290  callout function pointed to by <b>pcre_c
290  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
291  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.
292  </P>  </P>
293  <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>
294    <P>
295    The compiled form of a regular expression can be saved and re-used at a later
296    time, possibly by a different program, and even on a host other than the one on
297    which it was compiled. Details are given in the
298    <a href="pcreprecompile.html"><b>pcreprecompile</b></a>
299    documentation. However, compiling a regular expression with one version of PCRE
300    for use with a different version is not guaranteed to work and may cause
301    crashes.
302    </P>
303    <br><a name="SEC6" href="#TOC1">CHECKING BUILD-TIME OPTIONS</a><br>
304  <P>  <P>
305  <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>
306  </P>  </P>
# Line 203  documentation has more details about the Line 314  documentation has more details about the
314  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
315  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
316  which the information is placed. The following information is available:  which the information is placed. The following information is available:
 </P>  
 <P>  
317  <pre>  <pre>
318    PCRE_CONFIG_UTF8    PCRE_CONFIG_UTF8
319  </PRE>  </pre>
 </P>  
 <P>  
320  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;
321  otherwise it is set to zero.  otherwise it is set to zero.
322  </P>  <pre>
323  <P>    PCRE_CONFIG_UNICODE_PROPERTIES
324    </pre>
325    The output is an integer that is set to one if support for Unicode character
326    properties is available; otherwise it is set to zero.
327  <pre>  <pre>
328    PCRE_CONFIG_NEWLINE    PCRE_CONFIG_NEWLINE
329  </PRE>  </pre>
330  </P>  The output is an integer whose value specifies the default character sequence
331  <P>  that is recognized as meaning "newline". The four values that are supported
332  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
333  the newline character. It is either linefeed (10) or carriage return (13), and  default should normally be the standard sequence for your operating system.
 should normally be the standard character for your operating system.  
 </P>  
 <P>  
334  <pre>  <pre>
335    PCRE_CONFIG_LINK_SIZE    PCRE_CONFIG_LINK_SIZE
336  </PRE>  </pre>
 </P>  
 <P>  
337  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
338  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
339  allow larger regular expressions to be compiled, at the expense of slower  allow larger regular expressions to be compiled, at the expense of slower
340  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
341  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>  
342  <pre>  <pre>
343    PCRE_CONFIG_POSIX_MALLOC_THRESHOLD    PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
344  </PRE>  </pre>
 </P>  
 <P>  
345  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
346  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
347  the <b>pcreposix</b> documentation.  the
348  </P>  <a href="pcreposix.html"><b>pcreposix</b></a>
349  <P>  documentation.
350  <pre>  <pre>
351    PCRE_CONFIG_MATCH_LIMIT    PCRE_CONFIG_MATCH_LIMIT
352  </PRE>  </pre>
 </P>  
 <P>  
353  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
354  internal matching function calls in a <b>pcre_exec()</b> execution. Further  internal matching function calls in a <b>pcre_exec()</b> execution. Further
355  details are given with <b>pcre_exec()</b> below.  details are given with <b>pcre_exec()</b> below.
356  </P>  <pre>
357  <P>    PCRE_CONFIG_MATCH_LIMIT_RECURSION
358    </pre>
359    The output is an integer that gives the default limit for the depth of
360    recursion when calling the internal matching function in a <b>pcre_exec()</b>
361    execution. Further details are given with <b>pcre_exec()</b> below.
362  <pre>  <pre>
363    PCRE_CONFIG_STACKRECURSE    PCRE_CONFIG_STACKRECURSE
364  </PRE>  </pre>
365    The output is an integer that is set to one if internal recursion when running
366    <b>pcre_exec()</b> is implemented by recursive function calls that use the stack
367    to remember their state. This is the usual way that PCRE is compiled. The
368    output is zero if PCRE was compiled to use blocks of data on the heap instead
369    of recursive function calls. In this case, <b>pcre_stack_malloc</b> and
370    <b>pcre_stack_free</b> are called to manage memory blocks on the heap, thus
371    avoiding the use of the stack.
372  </P>  </P>
373  <P>  <br><a name="SEC7" href="#TOC1">COMPILING A PATTERN</a><br>
 The output is an integer that is set to one if internal recursion is  
 implemented by recursive function calls that use the stack to remember their  
 state. This is the usual way that PCRE is compiled. The output is zero if PCRE  
 was compiled to use blocks of data on the heap instead of recursive function  
 calls. In this case, <b>pcre_stack_malloc</b> and <b>pcre_stack_free</b> are  
 called to manage memory blocks on the heap, thus avoiding the use of the stack.  
 </P>  
 <br><a name="SEC5" href="#TOC1">COMPILING A PATTERN</a><br>  
374  <P>  <P>
375  <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>
376  <b>const char **<i>errptr</i>, int *<i>erroffset</i>,</b>  <b>const char **<i>errptr</i>, int *<i>erroffset</i>,</b>
377  <b>const unsigned char *<i>tableptr</i>);</b>  <b>const unsigned char *<i>tableptr</i>);</b>
378    <b>pcre *pcre_compile2(const char *<i>pattern</i>, int <i>options</i>,</b>
379    <b>int *<i>errorcodeptr</i>,</b>
380    <b>const char **<i>errptr</i>, int *<i>erroffset</i>,</b>
381    <b>const unsigned char *<i>tableptr</i>);</b>
382    </P>
383    <P>
384    Either of the functions <b>pcre_compile()</b> or <b>pcre_compile2()</b> can be
385    called to compile a pattern into an internal form. The only difference between
386    the two interfaces is that <b>pcre_compile2()</b> has an additional argument,
387    <i>errorcodeptr</i>, via which a numerical error code can be returned.
388  </P>  </P>
389  <P>  <P>
390  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
391  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
392  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
393  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
394  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
395  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.  
396  </P>  </P>
397  <P>  <P>
398  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
399  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
400  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>
401  which is an address (see below).  argument, which is an address (see below).
402  </P>  </P>
403  <P>  <P>
404  The <i>options</i> argument contains independent bits that affect the  The <i>options</i> argument contains various bit settings that affect the
405  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
406  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
407  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
408  in the <b>pcrepattern</b> documentation). For these options, the contents of the  the detailed description in the
409  <i>options</i> argument specifies their initial settings at the start of  <a href="pcrepattern.html"><b>pcrepattern</b></a>
410  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
411    specifies their initial settings at the start of compilation and execution. The
412    PCRE_ANCHORED and PCRE_NEWLINE_<i>xxx</i> options can be set at the time of
413  matching as well as at compile time.  matching as well as at compile time.
414  </P>  </P>
415  <P>  <P>
416  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.
417  Otherwise, if compilation of a pattern fails, <b>pcre_compile()</b> returns  Otherwise, if compilation of a pattern fails, <b>pcre_compile()</b> returns
418  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
419  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
420  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
421    where the error was discovered is placed in the variable pointed to by
422  <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.
423  </P>  </P>
424  <P>  <P>
425  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
426  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
427  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
428  <b>pcre_maketables()</b>. See the section on locale support below.  textual error message. Error codes and messages are listed below.
429  </P>  </P>
430  <P>  <P>
431  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
432    character tables that are built when PCRE is compiled, using the default C
433    locale. Otherwise, <i>tableptr</i> must be an address that is the result of a
434    call to <b>pcre_maketables()</b>. This value is stored with the compiled
435    pattern, and used again by <b>pcre_exec()</b>, unless another table pointer is
436    passed to it. For more discussion, see the section on locale support below.
437  </P>  </P>
438  <P>  <P>
439    This code fragment shows a typical straightforward call to <b>pcre_compile()</b>:
440  <pre>  <pre>
441    pcre *re;    pcre *re;
442    const char *error;    const char *error;
# Line 327  This code fragment shows a typical strai Line 447  This code fragment shows a typical strai
447      &error,           /* for error message */      &error,           /* for error message */
448      &erroffset,       /* for error offset */      &erroffset,       /* for error offset */
449      NULL);            /* use default character tables */      NULL);            /* use default character tables */
450  </PRE>  </pre>
451  </P>  The following names for option bits are defined in the <b>pcre.h</b> header
452  <P>  file:
 The following option bits are defined:  
 </P>  
 <P>  
453  <pre>  <pre>
454    PCRE_ANCHORED    PCRE_ANCHORED
455  </PRE>  </pre>
 </P>  
 <P>  
456  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
457  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
458  being searched (the "subject string"). This effect can also be achieved by  being searched (the "subject string"). This effect can also be achieved by
459  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
460  Perl.  Perl.
461  </P>  <pre>
462  <P>    PCRE_AUTO_CALLOUT
463    </pre>
464    If this bit is set, <b>pcre_compile()</b> automatically inserts callout items,
465    all with number 255, before each pattern item. For discussion of the callout
466    facility, see the
467    <a href="pcrecallout.html"><b>pcrecallout</b></a>
468    documentation.
469  <pre>  <pre>
470    PCRE_CASELESS    PCRE_CASELESS
471  </PRE>  </pre>
 </P>  
 <P>  
472  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
473  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
474  pattern by a (?i) option setting.  pattern by a (?i) option setting. In UTF-8 mode, PCRE always understands the
475  </P>  concept of case for characters whose values are less than 128, so caseless
476  <P>  matching is always possible. For characters with higher values, the concept of
477    case is supported if PCRE is compiled with Unicode property support, but not
478    otherwise. If you want to use caseless matching for characters 128 and above,
479    you must ensure that PCRE is compiled with Unicode property support as well as
480    with UTF-8 support.
481  <pre>  <pre>
482    PCRE_DOLLAR_ENDONLY    PCRE_DOLLAR_ENDONLY
483  </PRE>  </pre>
 </P>  
 <P>  
484  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
485  end of the subject string. Without this option, a dollar also matches  end of the subject string. Without this option, a dollar also matches
486  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
487  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.
488  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
489  a pattern.  pattern.
 </P>  
 <P>  
490  <pre>  <pre>
491    PCRE_DOTALL    PCRE_DOTALL
492  </PRE>  </pre>
 </P>  
 <P>  
493  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,
494  including newlines. Without it, newlines are excluded. This option is  including those that indicate newline. Without it, a dot does not match when
495  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
496  (?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
497  character, independent of the setting of this option.  negative class such as [^a] always matches newline characters, independent of
498  </P>  the setting of this option.
499  <P>  <pre>
500      PCRE_DUPNAMES
501    </pre>
502    If this bit is set, names used to identify capturing subpatterns need not be
503    unique. This can be helpful for certain types of pattern when it is known that
504    only one instance of the named subpattern can ever be matched. There are more
505    details of named subpatterns below; see also the
506    <a href="pcrepattern.html"><b>pcrepattern</b></a>
507    documentation.
508  <pre>  <pre>
509    PCRE_EXTENDED    PCRE_EXTENDED
510  </PRE>  </pre>
 </P>  
 <P>  
511  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
512  ignored except when escaped or inside a character class. Whitespace does not  ignored except when escaped or inside a character class. Whitespace does not
513  include the VT character (code 11). In addition, characters between an  include the VT character (code 11). In addition, characters between an
514  unescaped # outside a character class and the next newline character,  unescaped # outside a character class and the next newline, inclusive, are also
515  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
516  be changed within a pattern by a (?x) option setting.  pattern by a (?x) option setting.
517  </P>  </P>
518  <P>  <P>
519  This option makes it possible to include comments inside complicated patterns.  This option makes it possible to include comments inside complicated patterns.
520  Note, however, that this applies only to data characters. Whitespace characters  Note, however, that this applies only to data characters. Whitespace characters
521  may never appear within special character sequences in a pattern, for example  may never appear within special character sequences in a pattern, for example
522  within the sequence (?( which introduces a conditional subpattern.  within the sequence (?( which introduces a conditional subpattern.
 </P>  
 <P>  
523  <pre>  <pre>
524    PCRE_EXTRA    PCRE_EXTRA
525  </PRE>  </pre>
 </P>  
 <P>  
526  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
527  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
528  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
529  special meaning causes an error, thus reserving these combinations for future  special meaning causes an error, thus reserving these combinations for future
530  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
531  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
532  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
533  pattern.  this option. It can also be set by a (?X) option setting within a pattern.
534  </P>  <pre>
535  <P>    PCRE_FIRSTLINE
536    </pre>
537    If this option is set, an unanchored pattern is required to match before or at
538    the first newline in the subject string, though the matched text may continue
539    over the newline.
540  <pre>  <pre>
541    PCRE_MULTILINE    PCRE_MULTILINE
542  </PRE>  </pre>
543  </P>  By default, PCRE treats the subject string as consisting of a single line of
544  <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"  
545  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
546  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
547  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 428  Perl. Line 549  Perl.
549  </P>  </P>
550  <P>  <P>
551  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
552  match immediately following or immediately before any newline in the subject  match immediately following or immediately before internal newlines in the
553  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
554  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
555  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
556  occurrences of ^ or $ in a pattern, setting PCRE_MULTILINE has no effect.  occurrences of ^ or $ in a pattern, setting PCRE_MULTILINE has no effect.
557    <pre>
558      PCRE_NEWLINE_CR
559      PCRE_NEWLINE_LF
560      PCRE_NEWLINE_CRLF
561      PCRE_NEWLINE_ANYCRLF
562      PCRE_NEWLINE_ANY
563    </pre>
564    These options override the default newline definition that was chosen when PCRE
565    was built. Setting the first or the second specifies that a newline is
566    indicated by a single character (CR or LF, respectively). Setting
567    PCRE_NEWLINE_CRLF specifies that a newline is indicated by the two-character
568    CRLF sequence. Setting PCRE_NEWLINE_ANYCRLF specifies that any of the three
569    preceding sequences should be recognized. Setting PCRE_NEWLINE_ANY specifies
570    that any Unicode newline sequence should be recognized. The Unicode newline
571    sequences are the three just mentioned, plus the single characters VT (vertical
572    tab, U+000B), FF (formfeed, U+000C), NEL (next line, U+0085), LS (line
573    separator, U+2028), and PS (paragraph separator, U+2029). The last two are
574    recognized only in UTF-8 mode.
575    </P>
576    <P>
577    The newline setting in the options word uses three bits that are treated
578    as a number, giving eight possibilities. Currently only six are used (default
579    plus the five values above). This means that if you set more than one newline
580    option, the combination may or may not be sensible. For example,
581    PCRE_NEWLINE_CR with PCRE_NEWLINE_LF is equivalent to PCRE_NEWLINE_CRLF, but
582    other combinations may yield unused numbers and cause an error.
583    </P>
584    <P>
585    The only time that a line break is specially recognized when compiling a
586    pattern is if PCRE_EXTENDED is set, and an unescaped # outside a character
587    class is encountered. This indicates a comment that lasts until after the next
588    line break sequence. In other circumstances, line break sequences are treated
589    as literal data, except that in PCRE_EXTENDED mode, both CR and LF are treated
590    as whitespace characters and are therefore ignored.
591  </P>  </P>
592  <P>  <P>
593    The newline option that is set at compile time becomes the default that is used
594    for <b>pcre_exec()</b> and <b>pcre_dfa_exec()</b>, but it can be overridden.
595  <pre>  <pre>
596    PCRE_NO_AUTO_CAPTURE    PCRE_NO_AUTO_CAPTURE
597  </PRE>  </pre>
 </P>  
 <P>  
598  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
599  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
600  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
601  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
602  in Perl.  in Perl.
 </P>  
 <P>  
603  <pre>  <pre>
604    PCRE_UNGREEDY    PCRE_UNGREEDY
605  </PRE>  </pre>
 </P>  
 <P>  
606  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
607  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
608  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>  
609  <pre>  <pre>
610    PCRE_UTF8    PCRE_UTF8
611  </PRE>  </pre>
 </P>  
 <P>  
612  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
613  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
614  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
615  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
616  behaviour of PCRE are given in the  behaviour of PCRE are given in the
617  <a href="pcre.html#utf8support">section on UTF-8 support</a>  <a href="pcre.html#utf8support">section on UTF-8 support</a>
618  in the main  in the main
619  <a href="pcre.html"><b>pcre</b></a>  <a href="pcre.html"><b>pcre</b></a>
620  page.  page.
 </P>  
 <P>  
621  <pre>  <pre>
622    PCRE_NO_UTF8_CHECK    PCRE_NO_UTF8_CHECK
623  </PRE>  </pre>
 </P>  
 <P>  
624  When PCRE_UTF8 is set, the validity of the pattern as a UTF-8 string is  When PCRE_UTF8 is set, the validity of the pattern as a UTF-8 string is
625  automatically checked. If an invalid UTF-8 sequence of bytes is found,  automatically checked. There is a discussion about the
626  <b>pcre_compile()</b> returns an error. If you already know that your pattern is  <a href="pcre.html#utf8strings">validity of UTF-8 strings</a>
627  valid, and you want to skip this check for performance reasons, you can set the  in the main
628  PCRE_NO_UTF8_CHECK option. When it is set, the effect of passing an invalid  <a href="pcre.html"><b>pcre</b></a>
629  UTF-8 string as a pattern is undefined. It may cause your program to crash.  page. If an invalid UTF-8 sequence of bytes is found, <b>pcre_compile()</b>
630  Note that there is a similar option for suppressing the checking of subject  returns an error. If you already know that your pattern is valid, and you want
631  strings passed to <b>pcre_exec()</b>.  to skip this check for performance reasons, you can set the PCRE_NO_UTF8_CHECK
632    option. When it is set, the effect of passing an invalid UTF-8 string as a
633    pattern is undefined. It may cause your program to crash. Note that this option
634    can also be passed to <b>pcre_exec()</b> and <b>pcre_dfa_exec()</b>, to suppress
635    the UTF-8 validity checking of subject strings.
636    </P>
637    <br><a name="SEC8" href="#TOC1">COMPILATION ERROR CODES</a><br>
638    <P>
639    The following table lists the error codes than may be returned by
640    <b>pcre_compile2()</b>, along with the error messages that may be returned by
641    both compiling functions. As PCRE has developed, some error codes have fallen
642    out of use. To avoid confusion, they have not been re-used.
643    <pre>
644       0  no error
645       1  \ at end of pattern
646       2  \c at end of pattern
647       3  unrecognized character follows \
648       4  numbers out of order in {} quantifier
649       5  number too big in {} quantifier
650       6  missing terminating ] for character class
651       7  invalid escape sequence in character class
652       8  range out of order in character class
653       9  nothing to repeat
654      10  [this code is not in use]
655      11  internal error: unexpected repeat
656      12  unrecognized character after (?
657      13  POSIX named classes are supported only within a class
658      14  missing )
659      15  reference to non-existent subpattern
660      16  erroffset passed as NULL
661      17  unknown option bit(s) set
662      18  missing ) after comment
663      19  [this code is not in use]
664      20  regular expression too large
665      21  failed to get memory
666      22  unmatched parentheses
667      23  internal error: code overflow
668      24  unrecognized character after (?&#60;
669      25  lookbehind assertion is not fixed length
670      26  malformed number or name after (?(
671      27  conditional group contains more than two branches
672      28  assertion expected after (?(
673      29  (?R or (?[+-]digits must be followed by )
674      30  unknown POSIX class name
675      31  POSIX collating elements are not supported
676      32  this version of PCRE is not compiled with PCRE_UTF8 support
677      33  [this code is not in use]
678      34  character value in \x{...} sequence is too large
679      35  invalid condition (?(0)
680      36  \C not allowed in lookbehind assertion
681      37  PCRE does not support \L, \l, \N, \U, or \u
682      38  number after (?C is &#62; 255
683      39  closing ) for (?C expected
684      40  recursive call could loop indefinitely
685      41  unrecognized character after (?P
686      42  syntax error in subpattern name (missing terminator)
687      43  two named subpatterns have the same name
688      44  invalid UTF-8 string
689      45  support for \P, \p, and \X has not been compiled
690      46  malformed \P or \p sequence
691      47  unknown property name after \P or \p
692      48  subpattern name is too long (maximum 32 characters)
693      49  too many named subpatterns (maximum 10,000)
694      50  [this code is not in use]
695      51  octal value is greater than \377 (not in UTF-8 mode)
696      52  internal error: overran compiling workspace
697      53  internal error: previously-checked referenced subpattern not found
698      54  DEFINE group contains more than one branch
699      55  repeating a DEFINE group is not allowed
700      56  inconsistent NEWLINE options"
701      57  \g is not followed by a braced name or an optionally braced
702            non-zero number
703      58  (?+ or (?- or (?(+ or (?(- must be followed by a non-zero number
704    </PRE>
705  </P>  </P>
706  <br><a name="SEC6" href="#TOC1">STUDYING A PATTERN</a><br>  <br><a name="SEC9" href="#TOC1">STUDYING A PATTERN</a><br>
707  <P>  <P>
708  <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>
709  <b>const char **<i>errptr</i>);</b>  <b>const char **<i>errptr</i>);</b>
710  </P>  </P>
711  <P>  <P>
712  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
713  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
714  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
715  argument. If studing the pattern produces additional information that will help  argument. If studying the pattern produces additional information that will
716  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
717  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
718    results of the study.
719  </P>  </P>
720  <P>  <P>
721  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
722  <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
723  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
724  described below. If studying the pattern does not produce any additional  described
725  information, <b>pcre_study()</b> returns NULL. In that circumstance, if the  <a href="#extradata">below</a>
726  calling program wants to pass some of the other fields to <b>pcre_exec()</b>, it  in the section on matching a pattern.
727  must set up its own <b>pcre_extra</b> block.  </P>
728    <P>
729    If studying the pattern does not produce any additional information
730    <b>pcre_study()</b> returns NULL. In that circumstance, if the calling program
731    wants to pass any of the other fields to <b>pcre_exec()</b>, it must set up its
732    own <b>pcre_extra</b> block.
733  </P>  </P>
734  <P>  <P>
735  The second argument contains option bits. At present, no options are defined  The second argument of <b>pcre_study()</b> contains option bits. At present, no
736  for <b>pcre_study()</b>, and this argument should always be zero.  options are defined, and this argument should always be zero.
737  </P>  </P>
738  <P>  <P>
739  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
740  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
741  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
742  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
743  be sure that it has run successfully.  should test the error pointer for NULL after calling <b>pcre_study()</b>, to be
744    sure that it has run successfully.
745  </P>  </P>
746  <P>  <P>
747  This is a typical call to <b>pcre_study</b>():  This is a typical call to <b>pcre_study</b>():
 </P>  
 <P>  
748  <pre>  <pre>
749    pcre_extra *pe;    pcre_extra *pe;
750    pe = pcre_study(    pe = pcre_study(
751      re,             /* result of pcre_compile() */      re,             /* result of pcre_compile() */
752      0,              /* no options exist */      0,              /* no options exist */
753      &error);        /* set to NULL or points to a message */      &error);        /* set to NULL or points to a message */
754  </PRE>  </pre>
 </P>  
 <P>  
755  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
756  not have a single fixed starting character. A bitmap of possible starting  not have a single fixed starting character. A bitmap of possible starting
757  characters is created.  bytes is created.
758  </P>  <a name="localesupport"></a></P>
759  <a name="localesupport"></a><br><a name="SEC7" href="#TOC1">LOCALE SUPPORT</a><br>  <br><a name="SEC10" href="#TOC1">LOCALE SUPPORT</a><br>
760  <P>  <P>
761  PCRE handles caseless matching, and determines whether characters are letters,  PCRE handles caseless matching, and determines whether characters are letters,
762  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
763  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
764  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
765  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
766  is NULL, and is sufficient for many applications.  support. The use of locales with Unicode is discouraged. If you are handling
767  </P>  characters with codes greater than 128, you should either use UTF-8 and
768  <P>  Unicode, or use locales, but not try to mix the two.
769  An alternative set of tables can, however, be supplied. Such tables are built  </P>
770  by calling the <b>pcre_maketables()</b> function, which has no arguments, in the  <P>
771  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
772  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.
773  French locale (where accented characters with codes greater than 128 are  Normally, the internal tables recognize only ASCII characters. However, when
774  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
775  </P>  default "C" locale of the local system, which may cause them to be different.
776  <P>  </P>
777    <P>
778    The internal tables can always be overridden by tables supplied by the
779    application that calls PCRE. These may be created in a different locale from
780    the default. As more and more applications change to using Unicode, the need
781    for this locale support is expected to die away.
782    </P>
783    <P>
784    External tables are built by calling the <b>pcre_maketables()</b> function,
785    which has no arguments, in the relevant locale. The result can then be passed
786    to <b>pcre_compile()</b> or <b>pcre_exec()</b> as often as necessary. For
787    example, to build and use tables that are appropriate for the French locale
788    (where accented characters with values greater than 128 are treated as letters),
789    the following code could be used:
790  <pre>  <pre>
791    setlocale(LC_CTYPE, "fr");    setlocale(LC_CTYPE, "fr_FR");
792    tables = pcre_maketables();    tables = pcre_maketables();
793    re = pcre_compile(..., tables);    re = pcre_compile(..., tables);
794  </PRE>  </pre>
795    The locale name "fr_FR" is used on Linux and other Unix-like systems; if you
796    are using Windows, the name for the French locale is "french".
797  </P>  </P>
798  <P>  <P>
799  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
800  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
801    that the memory containing the tables remains available for as long as it is
802    needed.
803    </P>
804    <P>
805    The pointer that is passed to <b>pcre_compile()</b> is saved with the compiled
806  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>
807  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
808  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
809  in different locales. It is the caller's responsibility to ensure that the  different patterns can be compiled in different locales.
810  memory containing the tables remains available for as long as it is needed.  </P>
811    <P>
812    It is possible to pass a table pointer or NULL (indicating the use of the
813    internal tables) to <b>pcre_exec()</b>. Although not intended for this purpose,
814    this facility could be used to match a pattern in a different locale from the
815    one in which it was compiled. Passing table pointers at run time is discussed
816    below in the section on matching a pattern.
817  </P>  </P>
818  <br><a name="SEC8" href="#TOC1">INFORMATION ABOUT A PATTERN</a><br>  <br><a name="SEC11" href="#TOC1">INFORMATION ABOUT A PATTERN</a><br>
819  <P>  <P>
820  <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>
821  <b>int <i>what</i>, void *<i>where</i>);</b>  <b>int <i>what</i>, void *<i>where</i>);</b>
# Line 587  the pattern was not studied. The third a Line 832  the pattern was not studied. The third a
832  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
833  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
834  the following negative numbers:  the following negative numbers:
 </P>  
 <P>  
835  <pre>  <pre>
836    PCRE_ERROR_NULL       the argument <i>code</i> was NULL    PCRE_ERROR_NULL       the argument <i>code</i> was NULL
837                          the argument <i>where</i> was NULL                          the argument <i>where</i> was NULL
838    PCRE_ERROR_BADMAGIC   the "magic number" was not found    PCRE_ERROR_BADMAGIC   the "magic number" was not found
839    PCRE_ERROR_BADOPTION  the value of <i>what</i> was invalid    PCRE_ERROR_BADOPTION  the value of <i>what</i> was invalid
840  </PRE>  </pre>
841  </P>  The "magic number" is placed at the start of each compiled pattern as an simple
842  <P>  check against passing an arbitrary memory pointer. Here is a typical call of
843  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>  
844  <pre>  <pre>
845    int rc;    int rc;
846    unsigned long int length;    size_t length;
847    rc = pcre_fullinfo(    rc = pcre_fullinfo(
848      re,               /* result of pcre_compile() */      re,               /* result of pcre_compile() */
849      pe,               /* result of pcre_study(), or NULL */      pe,               /* result of pcre_study(), or NULL */
850      PCRE_INFO_SIZE,   /* what is required */      PCRE_INFO_SIZE,   /* what is required */
851      &length);         /* where to put the data */      &length);         /* where to put the data */
852  </PRE>  </pre>
 </P>  
 <P>  
853  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
854  as follows:  as follows:
 </P>  
 <P>  
855  <pre>  <pre>
856    PCRE_INFO_BACKREFMAX    PCRE_INFO_BACKREFMAX
857  </PRE>  </pre>
 </P>  
 <P>  
858  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
859  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
860  no back references.  no back references.
 </P>  
 <P>  
861  <pre>  <pre>
862    PCRE_INFO_CAPTURECOUNT    PCRE_INFO_CAPTURECOUNT
863  </PRE>  </pre>
 </P>  
 <P>  
864  Return the number of capturing subpatterns in the pattern. The fourth argument  Return the number of capturing subpatterns in the pattern. The fourth argument
865  should point to an \fbint\fR variable.  should point to an <b>int</b> variable.
866  </P>  <pre>
867  <P>    PCRE_INFO_DEFAULT_TABLES
868    </pre>
869    Return a pointer to the internal default character tables within PCRE. The
870    fourth argument should point to an <b>unsigned char *</b> variable. This
871    information call is provided for internal use by the <b>pcre_study()</b>
872    function. External callers can cause PCRE to use its internal tables by passing
873    a NULL table pointer.
874  <pre>  <pre>
875    PCRE_INFO_FIRSTBYTE    PCRE_INFO_FIRSTBYTE
876  </PRE>  </pre>
 </P>  
 <P>  
877  Return information about the first byte of any matched string, for a  Return information about the first byte of any matched string, for a
878  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>
879  old name is still recognized for backwards compatibility.)  variable. (This option used to be called PCRE_INFO_FIRSTCHAR; the old name is
880  </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  
881  </P>  </P>
882  <P>  <P>
883    If there is a fixed first byte, for example, from a pattern such as
884    (cat|cow|coyote), its value is returned. Otherwise, if either
885    <br>
886    <br>
887  (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
888  starts with "^", or  starts with "^", or
889  </P>  <br>
890  <P>  <br>
891  (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
892  (if it were set, the pattern would be anchored),  (if it were set, the pattern would be anchored),
893  </P>  <br>
894  <P>  <br>
895  -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
896  subject string or after any newline within the string. Otherwise -2 is  subject string or after any newline within the string. Otherwise -2 is
897  returned. For anchored patterns, -2 is returned.  returned. For anchored patterns, -2 is returned.
 </P>  
 <P>  
898  <pre>  <pre>
899    PCRE_INFO_FIRSTTABLE    PCRE_INFO_FIRSTTABLE
900  </PRE>  </pre>
 </P>  
 <P>  
901  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
902  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
903  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
904  fourth argument should point to an <b>unsigned char *</b> variable.  fourth argument should point to an <b>unsigned char *</b> variable.
905  </P>  <pre>
906  <P>    PCRE_INFO_HASCRORLF
907    </pre>
908    Return 1 if the pattern contains any explicit matches for CR or LF characters,
909    otherwise 0. The fourth argument should point to an <b>int</b> variable.
910    <pre>
911      PCRE_INFO_JCHANGED
912    </pre>
913    Return 1 if the (?J) option setting is used in the pattern, otherwise 0. The
914    fourth argument should point to an <b>int</b> variable. The (?J) internal option
915    setting changes the local PCRE_DUPNAMES option.
916  <pre>  <pre>
917    PCRE_INFO_LASTLITERAL    PCRE_INFO_LASTLITERAL
918  </PRE>  </pre>
 </P>  
 <P>  
919  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
920  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
921  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 685  returned. For anchored patterns, a last Line 923  returned. For anchored patterns, a last
923  follows something of variable length. For example, for the pattern  follows something of variable length. For example, for the pattern
924  /^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
925  is -1.  is -1.
 </P>  
 <P>  
926  <pre>  <pre>
927    PCRE_INFO_NAMECOUNT    PCRE_INFO_NAMECOUNT
928    PCRE_INFO_NAMEENTRYSIZE    PCRE_INFO_NAMEENTRYSIZE
929    PCRE_INFO_NAMETABLE    PCRE_INFO_NAMETABLE
930  </PRE>  </pre>
 </P>  
 <P>  
931  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
932  names are just an additional way of identifying the parentheses, which still  names are just an additional way of identifying the parentheses, which still
933  acquire a number. A caller that wants to extract data from a named subpattern  acquire numbers. Several convenience functions such as
934  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
935  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
936  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
937  table for the pattern.  output vector (described with <b>pcre_exec()</b> below). To do the conversion,
938    you need to use the name-to-number map, which is described by these three
939    values.
940  </P>  </P>
941  <P>  <P>
942  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 710  length of the longest name. PCRE_INFO_NA Line 946  length of the longest name. PCRE_INFO_NA
946  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
947  are the number of the capturing parenthesis, most significant byte first. The  are the number of the capturing parenthesis, most significant byte first. The
948  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
949  alphabetical order. For example, consider the following pattern (assume  alphabetical order. When PCRE_DUPNAMES is set, duplicate names are in order of
950    their parentheses numbers. For example, consider the following pattern (assume
951  PCRE_EXTENDED is set, so white space - including newlines - is ignored):  PCRE_EXTENDED is set, so white space - including newlines - is ignored):
 </P>  
 <P>  
952  <pre>  <pre>
953    (?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) )
954    (?P&#60;month&#62;\d\d) - (?P&#60;day&#62;\d\d) )  </pre>
 </PRE>  
 </P>  
 <P>  
955  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
956  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
957  bytes shows in hex, and undefined bytes shown as ??:  bytes shows in hexadecimal, and undefined bytes shown as ??:
 </P>  
 <P>  
958  <pre>  <pre>
959    00 01 d  a  t  e  00 ??    00 01 d  a  t  e  00 ??
960    00 05 d  a  y  00 ?? ??    00 05 d  a  y  00 ?? ??
961    00 04 m  o  n  t  h  00    00 04 m  o  n  t  h  00
962    00 02 y  e  a  r  00 ??    00 02 y  e  a  r  00 ??
963  </PRE>  </pre>
964  </P>  When writing code to extract data from named subpatterns using the
965  <P>  name-to-number map, remember that the length of the entries is likely to be
966  When writing code to extract data from named subpatterns, remember that the  different for each compiled pattern.
967  length of each entry may be different for each compiled pattern.  <pre>
968  </P>    PCRE_INFO_OKPARTIAL
969  <P>  </pre>
970    Return 1 if the pattern can be used for partial matching, otherwise 0. The
971    fourth argument should point to an <b>int</b> variable. The
972    <a href="pcrepartial.html"><b>pcrepartial</b></a>
973    documentation lists the restrictions that apply to patterns when partial
974    matching is used.
975  <pre>  <pre>
976    PCRE_INFO_OPTIONS    PCRE_INFO_OPTIONS
977  </PRE>  </pre>
 </P>  
 <P>  
978  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
979  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
980  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
981  top-level option settings within the pattern itself.  top-level option settings at the start of the pattern itself. In other words,
982    they are the options that will be in force when matching starts. For example,
983    if the pattern /(?im)abc(?-i)d/ is compiled with the PCRE_EXTENDED option, the
984    result is PCRE_CASELESS, PCRE_MULTILINE, and PCRE_EXTENDED.
985  </P>  </P>
986  <P>  <P>
987  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
988  alternatives begin with one of the following:  alternatives begin with one of the following:
 </P>  
 <P>  
989  <pre>  <pre>
990    ^     unless PCRE_MULTILINE is set    ^     unless PCRE_MULTILINE is set
991    \A    always    \A    always
992    \G    always    \G    always
993    .*    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
994            references to the subpattern in which .* appears  </pre>
 </PRE>  
 </P>  
 <P>  
995  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
996  <b>pcre_fullinfo()</b>.  <b>pcre_fullinfo()</b>.
 </P>  
 <P>  
997  <pre>  <pre>
998    PCRE_INFO_SIZE    PCRE_INFO_SIZE
999  </PRE>  </pre>
 </P>  
 <P>  
1000  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
1001  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
1002  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>
1003  variable.  variable.
 </P>  
 <P>  
1004  <pre>  <pre>
1005    PCRE_INFO_STUDYSIZE    PCRE_INFO_STUDYSIZE
1006  </PRE>  </pre>
1007  </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  
1008  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
1009  <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
1010  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
1011  <b>size_t</b> variable.  <b>size_t</b> variable.
1012  </P>  </P>
1013  <br><a name="SEC9" href="#TOC1">OBSOLETE INFO FUNCTION</a><br>  <br><a name="SEC12" href="#TOC1">OBSOLETE INFO FUNCTION</a><br>
1014  <P>  <P>
1015  <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>
1016  <b>*<i>firstcharptr</i>);</b>  <b>*<i>firstcharptr</i>);</b>
# Line 798  restrictive to return all the available Line 1021  restrictive to return all the available
1021  programs should use <b>pcre_fullinfo()</b> instead. The yield of  programs should use <b>pcre_fullinfo()</b> instead. The yield of
1022  <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
1023  following negative numbers:  following negative numbers:
 </P>  
 <P>  
1024  <pre>  <pre>
1025    PCRE_ERROR_NULL       the argument <i>code</i> was NULL    PCRE_ERROR_NULL       the argument <i>code</i> was NULL
1026    PCRE_ERROR_BADMAGIC   the "magic number" was not found    PCRE_ERROR_BADMAGIC   the "magic number" was not found
1027  </PRE>  </pre>
 </P>  
 <P>  
1028  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
1029  pattern was compiled is placed in the integer it points to (see  pattern was compiled is placed in the integer it points to (see
1030  PCRE_INFO_OPTIONS above).  PCRE_INFO_OPTIONS above).
# Line 815  If the pattern is not anchored and the < Line 1034  If the pattern is not anchored and the <
1034  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
1035  string (see PCRE_INFO_FIRSTBYTE above).  string (see PCRE_INFO_FIRSTBYTE above).
1036  </P>  </P>
1037  <br><a name="SEC10" href="#TOC1">MATCHING A PATTERN</a><br>  <br><a name="SEC13" href="#TOC1">REFERENCE COUNTS</a><br>
1038    <P>
1039    <b>int pcre_refcount(pcre *<i>code</i>, int <i>adjust</i>);</b>
1040    </P>
1041    <P>
1042    The <b>pcre_refcount()</b> function is used to maintain a reference count in the
1043    data block that contains a compiled pattern. It is provided for the benefit of
1044    applications that operate in an object-oriented manner, where different parts
1045    of the application may be using the same compiled pattern, but you want to free
1046    the block when they are all done.
1047    </P>
1048    <P>
1049    When a pattern is compiled, the reference count field is initialized to zero.
1050    It is changed only by calling this function, whose action is to add the
1051    <i>adjust</i> value (which may be positive or negative) to it. The yield of the
1052    function is the new value. However, the value of the count is constrained to
1053    lie between 0 and 65535, inclusive. If the new value is outside these limits,
1054    it is forced to the appropriate limit value.
1055    </P>
1056    <P>
1057    Except when it is zero, the reference count is not correctly preserved if a
1058    pattern is compiled on one host and then transferred to a host whose byte-order
1059    is different. (This seems a highly unlikely scenario.)
1060    </P>
1061    <br><a name="SEC14" href="#TOC1">MATCHING A PATTERN: THE TRADITIONAL FUNCTION</a><br>
1062  <P>  <P>
1063  <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>
1064  <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 823  string (see PCRE_INFO_FIRSTBYTE above). Line 1066  string (see PCRE_INFO_FIRSTBYTE above).
1066  </P>  </P>
1067  <P>  <P>
1068  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
1069  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
1070  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
1071  <i>extra</i> argument.  <i>extra</i> argument. This function is the main matching facility of the
1072    library, and it operates in a Perl-like manner. For specialist use there is
1073    also an alternative matching function, which is described
1074    <a href="#dfamatch">below</a>
1075    in the section about the <b>pcre_dfa_exec()</b> function.
1076  </P>  </P>
1077  <P>  <P>
1078  Here is an example of a simple call to <b>pcre_exec()</b>:  In most applications, the pattern will have been compiled (and optionally
1079    studied) in the same process that calls <b>pcre_exec()</b>. However, it is
1080    possible to save compiled patterns and study data, and then use them later
1081    in different processes, possibly even on different hosts. For a discussion
1082    about this, see the
1083    <a href="pcreprecompile.html"><b>pcreprecompile</b></a>
1084    documentation.
1085  </P>  </P>
1086  <P>  <P>
1087    Here is an example of a simple call to <b>pcre_exec()</b>:
1088  <pre>  <pre>
1089    int rc;    int rc;
1090    int ovector[30];    int ovector[30];
# Line 841  Here is an example of a simple call to < Line 1095  Here is an example of a simple call to <
1095      11,             /* the length of the subject string */      11,             /* the length of the subject string */
1096      0,              /* start at offset 0 in the subject */      0,              /* start at offset 0 in the subject */
1097      0,              /* default options */      0,              /* default options */
1098      ovector,        /* vector for substring information */      ovector,        /* vector of integers for substring information */
1099      30);            /* number of elements in the vector */      30);            /* number of elements (NOT size in bytes) */
1100  </PRE>  <a name="extradata"></a></PRE>
1101  </P>  </P>
1102    <br><b>
1103    Extra data for <b>pcre_exec()</b>
1104    </b><br>
1105  <P>  <P>
1106  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>
1107  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
1108  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
1109  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
1110  </P>  fields (not necessarily in this order):
 <P>  
1111  <pre>  <pre>
1112    unsigned long int <i>flags</i>;    unsigned long int <i>flags</i>;
1113    void *<i>study_data</i>;    void *<i>study_data</i>;
1114    unsigned long int <i>match_limit</i>;    unsigned long int <i>match_limit</i>;
1115      unsigned long int <i>match_limit_recursion</i>;
1116    void *<i>callout_data</i>;    void *<i>callout_data</i>;
1117  </PRE>    const unsigned char *<i>tables</i>;
1118  </P>  </pre>
 <P>  
1119  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
1120  are set. The flag bits are:  are set. The flag bits are:
 </P>  
 <P>  
1121  <pre>  <pre>
1122    PCRE_EXTRA_STUDY_DATA    PCRE_EXTRA_STUDY_DATA
1123    PCRE_EXTRA_MATCH_LIMIT    PCRE_EXTRA_MATCH_LIMIT
1124      PCRE_EXTRA_MATCH_LIMIT_RECURSION
1125    PCRE_EXTRA_CALLOUT_DATA    PCRE_EXTRA_CALLOUT_DATA
1126  </PRE>    PCRE_EXTRA_TABLES
1127  </P>  </pre>
 <P>  
1128  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
1129  <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
1130  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
1131  the block by setting the other fields.  the block by setting the other fields and their corresponding flag bits.
1132  </P>  </P>
1133  <P>  <P>
1134  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
1135  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,
1136  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
1137  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.  
1138  </P>  </P>
1139  <P>  <P>
1140  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
1141  default is 10 million, which handles all but the most extreme cases. You can  (sometimes recursively). The limit set by <i>match_limit</i> is imposed on the
1142  reduce the default by suppling <b>pcre_exec()</b> with a \fRpcre_extra\fR block  number of times this function is called during a match, which has the effect of
1143  in which <i>match_limit</i> is set to a smaller value, and  limiting the amount of backtracking that can take place. For patterns that are
1144  PCRE_EXTRA_MATCH_LIMIT is set in the <i>flags</i> field. If the limit is  not anchored, the count restarts from zero for each position in the subject
1145  exceeded, <b>pcre_exec()</b> returns PCRE_ERROR_MATCHLIMIT.  string.
1146  </P>  </P>
1147  <P>  <P>
1148  The <i>pcre_callout</i> field is used in conjunction with the "callout" feature,  The default value for the limit can be set when PCRE is built; the default
1149  which is described in the <b>pcrecallout</b> documentation.  default is 10 million, which handles all but the most extreme cases. You can
1150  </P>  override the default by suppling <b>pcre_exec()</b> with a <b>pcre_extra</b>
1151  <P>  block in which <i>match_limit</i> is set, and PCRE_EXTRA_MATCH_LIMIT is set in
1152  The PCRE_ANCHORED option can be passed in the <i>options</i> argument, whose  the <i>flags</i> field. If the limit is exceeded, <b>pcre_exec()</b> returns
1153  unused bits must be zero. This limits <b>pcre_exec()</b> to matching at the  PCRE_ERROR_MATCHLIMIT.
1154  first matching position. However, if a pattern was compiled with PCRE_ANCHORED,  </P>
1155  or turned out to be anchored by virtue of its contents, it cannot be made  <P>
1156  unachored at matching time.  The <i>match_limit_recursion</i> field is similar to <i>match_limit</i>, but
1157    instead of limiting the total number of times that <b>match()</b> is called, it
1158    limits the depth of recursion. The recursion depth is a smaller number than the
1159    total number of calls, because not all calls to <b>match()</b> are recursive.
1160    This limit is of use only if it is set smaller than <i>match_limit</i>.
1161    </P>
1162    <P>
1163    Limiting the recursion depth limits the amount of stack that can be used, or,
1164    when PCRE has been compiled to use memory on the heap instead of the stack, the
1165    amount of heap memory that can be used.
1166    </P>
1167    <P>
1168    The default value for <i>match_limit_recursion</i> can be set when PCRE is
1169    built; the default default is the same value as the default for
1170    <i>match_limit</i>. You can override the default by suppling <b>pcre_exec()</b>
1171    with a <b>pcre_extra</b> block in which <i>match_limit_recursion</i> is set, and
1172    PCRE_EXTRA_MATCH_LIMIT_RECURSION is set in the <i>flags</i> field. If the limit
1173    is exceeded, <b>pcre_exec()</b> returns PCRE_ERROR_RECURSIONLIMIT.
1174  </P>  </P>
1175  <P>  <P>
1176  When PCRE_UTF8 was set at compile time, the validity of the subject as a UTF-8  The <i>pcre_callout</i> field is used in conjunction with the "callout" feature,
1177  string is automatically checked, and the value of <i>startoffset</i> is also  which is described in the
1178  checked to ensure that it points to the start of a UTF-8 character. If an  <a href="pcrecallout.html"><b>pcrecallout</b></a>
1179  invalid UTF-8 sequence of bytes is found, <b>pcre_exec()</b> returns the error  documentation.
 PCRE_ERROR_BADUTF8. If <i>startoffset</i> contains an invalid value,  
 PCRE_ERROR_BADUTF8_OFFSET is returned.  
1180  </P>  </P>
1181  <P>  <P>
1182  If you already know that your subject is valid, and you want to skip these  The <i>tables</i> field is used to pass a character tables pointer to
1183  checks for performance reasons, you can set the PCRE_NO_UTF8_CHECK option when  <b>pcre_exec()</b>; this overrides the value that is stored with the compiled
1184  calling <b>pcre_exec()</b>. You might want to do this for the second and  pattern. A non-NULL value is stored with the compiled pattern only if custom
1185  subsequent calls to <b>pcre_exec()</b> if you are making repeated calls to find  tables were supplied to <b>pcre_compile()</b> via its <i>tableptr</i> argument.
1186  all the matches in a single subject string. However, you should be sure that  If NULL is passed to <b>pcre_exec()</b> using this mechanism, it forces PCRE's
1187  the value of <i>startoffset</i> points to the start of a UTF-8 character. When  internal tables to be used. This facility is helpful when re-using patterns
1188  PCRE_NO_UTF8_CHECK is set, the effect of passing an invalid UTF-8 string as a  that have been saved after compiling with an external set of tables, because
1189  subject, or a value of <i>startoffset</i> that does not point to the start of a  the external tables might be at a different address when <b>pcre_exec()</b> is
1190  UTF-8 character, is undefined. Your program may crash.  called. See the
1191    <a href="pcreprecompile.html"><b>pcreprecompile</b></a>
1192    documentation for a discussion of saving compiled patterns for later use.
1193    <a name="execoptions"></a></P>
1194    <br><b>
1195    Option bits for <b>pcre_exec()</b>
1196    </b><br>
1197    <P>
1198    The unused bits of the <i>options</i> argument for <b>pcre_exec()</b> must be
1199    zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_<i>xxx</i>,
1200    PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NO_UTF8_CHECK and PCRE_PARTIAL.
1201    <pre>
1202      PCRE_ANCHORED
1203    </pre>
1204    The PCRE_ANCHORED option limits <b>pcre_exec()</b> to matching at the first
1205    matching position. If a pattern was compiled with PCRE_ANCHORED, or turned out
1206    to be anchored by virtue of its contents, it cannot be made unachored at
1207    matching time.
1208    <pre>
1209      PCRE_NEWLINE_CR
1210      PCRE_NEWLINE_LF
1211      PCRE_NEWLINE_CRLF
1212      PCRE_NEWLINE_ANYCRLF
1213      PCRE_NEWLINE_ANY
1214    </pre>
1215    These options override the newline definition that was chosen or defaulted when
1216    the pattern was compiled. For details, see the description of
1217    <b>pcre_compile()</b> above. During matching, the newline choice affects the
1218    behaviour of the dot, circumflex, and dollar metacharacters. It may also alter
1219    the way the match position is advanced after a match failure for an unanchored
1220    pattern.
1221  </P>  </P>
1222  <P>  <P>
1223  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
1224    match attempt for an unanchored pattern fails when the current position is at a
1225    CRLF sequence, and the pattern contains no explicit matches for CR or NL
1226    characters, the match position is advanced by two characters instead of one, in
1227    other words, to after the CRLF.
1228    </P>
1229    <P>
1230    The above rule is a compromise that makes the most common cases work as
1231    expected. For example, if the pattern is .+A (and the PCRE_DOTALL option is not
1232    set), it does not match the string "\r\nA" because, after failing at the
1233    start, it skips both the CR and the LF before retrying. However, the pattern
1234    [\r\n]A does match that string, because it contains an explicit CR or LF
1235    reference, and so advances only by one character after the first failure.
1236    Note than an explicit CR or LF reference occurs for negated character classes
1237    such as [^X] because they can match CR or LF characters.
1238  </P>  </P>
1239  <P>  <P>
1240    Notwithstanding the above, anomalous effects may still occur when CRLF is a
1241    valid newline sequence and explicit \r or \n escapes appear in the pattern.
1242  <pre>  <pre>
1243    PCRE_NOTBOL    PCRE_NOTBOL
1244  </PRE>  </pre>
1245  </P>  This option specifies that first character of the subject string is not the
1246  <P>  beginning of a line, so the circumflex metacharacter should not match before
1247  The first character of the string is not the beginning of a line, so the  it. Setting this without PCRE_MULTILINE (at compile time) causes circumflex
1248  circumflex metacharacter should not match before it. Setting this without  never to match. This option affects only the behaviour of the circumflex
1249  PCRE_MULTILINE (at compile time) causes circumflex never to match.  metacharacter. It does not affect \A.
 </P>  
 <P>  
1250  <pre>  <pre>
1251    PCRE_NOTEOL    PCRE_NOTEOL
1252  </PRE>  </pre>
1253  </P>  This option specifies that the end of the subject string is not the end of a
1254  <P>  line, so the dollar metacharacter should not match it nor (except in multiline
1255  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
1256  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
1257  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>  
1258  <pre>  <pre>
1259    PCRE_NOTEMPTY    PCRE_NOTEMPTY
1260  </PRE>  </pre>
 </P>  
 <P>  
1261  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
1262  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
1263  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>  
1264  <pre>  <pre>
1265    a?b?    a?b?
1266  </PRE>  </pre>
 </P>  
 <P>  
1267  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
1268  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
1269  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 974  Perl has no direct equivalent of PCRE_NO Line 1273  Perl has no direct equivalent of PCRE_NO
1273  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
1274  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
1275  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
1276  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
1277  below) and trying an ordinary match again.  starting offset (see below) and trying an ordinary match again. There is some
1278    code that demonstrates how to do this in the <i>pcredemo.c</i> sample program.
1279    <pre>
1280      PCRE_NO_UTF8_CHECK
1281    </pre>
1282    When PCRE_UTF8 is set at compile time, the validity of the subject as a UTF-8
1283    string is automatically checked when <b>pcre_exec()</b> is subsequently called.
1284    The value of <i>startoffset</i> is also checked to ensure that it points to the
1285    start of a UTF-8 character. There is a discussion about the validity of UTF-8
1286    strings in the
1287    <a href="pcre.html#utf8strings">section on UTF-8 support</a>
1288    in the main
1289    <a href="pcre.html"><b>pcre</b></a>
1290    page. If an invalid UTF-8 sequence of bytes is found, <b>pcre_exec()</b> returns
1291    the error PCRE_ERROR_BADUTF8. If <i>startoffset</i> contains an invalid value,
1292    PCRE_ERROR_BADUTF8_OFFSET is returned.
1293  </P>  </P>
1294  <P>  <P>
1295  The subject string is passed to <b>pcre_exec()</b> as a pointer in  If you already know that your subject is valid, and you want to skip these
1296  <i>subject</i>, a length in <i>length</i>, and a starting byte offset in  checks for performance reasons, you can set the PCRE_NO_UTF8_CHECK option when
1297  <i>startoffset</i>. Unlike the pattern string, the subject may contain binary  calling <b>pcre_exec()</b>. You might want to do this for the second and
1298  zero bytes. When the starting offset is zero, the search for a match starts at  subsequent calls to <b>pcre_exec()</b> if you are making repeated calls to find
1299  the beginning of the subject, and this is by far the most common case.  all the matches in a single subject string. However, you should be sure that
1300    the value of <i>startoffset</i> points to the start of a UTF-8 character. When
1301    PCRE_NO_UTF8_CHECK is set, the effect of passing an invalid UTF-8 string as a
1302    subject, or a value of <i>startoffset</i> that does not point to the start of a
1303    UTF-8 character, is undefined. Your program may crash.
1304    <pre>
1305      PCRE_PARTIAL
1306    </pre>
1307    This option turns on the partial matching feature. If the subject string fails
1308    to match the pattern, but at some point during the matching process the end of
1309    the subject was reached (that is, the subject partially matches the pattern and
1310    the failure to match occurred only because there were not enough subject
1311    characters), <b>pcre_exec()</b> returns PCRE_ERROR_PARTIAL instead of
1312    PCRE_ERROR_NOMATCH. When PCRE_PARTIAL is used, there are restrictions on what
1313    may appear in the pattern. These are discussed in the
1314    <a href="pcrepartial.html"><b>pcrepartial</b></a>
1315    documentation.
1316  </P>  </P>
1317    <br><b>
1318    The string to be matched by <b>pcre_exec()</b>
1319    </b><br>
1320  <P>  <P>
1321  If the pattern was compiled with the PCRE_UTF8 option, the subject must be a  The subject string is passed to <b>pcre_exec()</b> as a pointer in
1322  sequence of bytes that is a valid UTF-8 string, and the starting offset must  <i>subject</i>, a length in <i>length</i>, and a starting byte offset in
1323  point to the beginning of a UTF-8 character. If an invalid UTF-8 string or  <i>startoffset</i>. In UTF-8 mode, the byte offset must point to the start of a
1324  offset is passed, an error (either PCRE_ERROR_BADUTF8 or  UTF-8 character. Unlike the pattern string, the subject may contain binary zero
1325  PCRE_ERROR_BADUTF8_OFFSET) is returned, unless the option PCRE_NO_UTF8_CHECK is  bytes. When the starting offset is zero, the search for a match starts at the
1326  set, in which case PCRE's behaviour is not defined.  beginning of the subject, and this is by far the most common case.
1327  </P>  </P>
1328  <P>  <P>
1329  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 998  same subject by calling <b>pcre_exec()</ Line 1331  same subject by calling <b>pcre_exec()</
1331  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
1332  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
1333  lookbehind. For example, consider the pattern  lookbehind. For example, consider the pattern
 </P>  
 <P>  
1334  <pre>  <pre>
1335    \Biss\B    \Biss\B
1336  </PRE>  </pre>
 </P>  
 <P>  
1337  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
1338  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
1339  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 1017  behind the starting point to discover th Line 1346  behind the starting point to discover th
1346  </P>  </P>
1347  <P>  <P>
1348  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
1349  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
1350  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.
1351  </P>  </P>
1352    <br><b>
1353    How <b>pcre_exec()</b> returns captured substrings
1354    </b><br>
1355  <P>  <P>
1356  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
1357  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 1031  kinds of parenthesized subpattern that d Line 1363  kinds of parenthesized subpattern that d
1363  <P>  <P>
1364  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
1365  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
1366  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>:
1367  back captured substrings, each substring using a pair of integers. The  this argument is NOT the size of <i>ovector</i> in bytes.
1368  remaining third of the vector is used as workspace by <b>pcre_exec()</b> while  </P>
1369  matching capturing subpatterns, and is not available for passing back  <P>
1370  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,
1371  three. If it is not, it is rounded down.  each substring using a pair of integers. The remaining third of the vector is
1372    used as workspace by <b>pcre_exec()</b> while matching capturing subpatterns,
1373    and is not available for passing back information. The length passed in
1374    <i>ovecsize</i> should always be a multiple of three. If it is not, it is
1375    rounded down.
1376  </P>  </P>
1377  <P>  <P>
1378  When a match has been successful, information about captured substrings is  When a match is successful, information about captured substrings is returned
1379  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
1380  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
1381  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
1382  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
1383  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
1384  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
1385  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>
1386  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
1387  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
1388  just the first pair of offsets has been set.  capturing subpatterns, the return value from a successful match is 1,
1389  </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.  
1390  </P>  </P>
1391  <P>  <P>
1392  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
1393  string that it matched that gets returned.  string that it matched that is returned.
1394  </P>  </P>
1395  <P>  <P>
1396  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
1397  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
1398  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
1399  <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
1400  <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
1401  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
1402  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
1403  to supply an <i>ovector</i>.  advisable to supply an <i>ovector</i>.
1404  </P>  </P>
1405  <P>  <P>
1406  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
1407  subpatterns there are in a compiled pattern. The smallest size for  subpatterns there are in a compiled pattern. The smallest size for
1408  <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
1409  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.
1410  </P>  </P>
1411  <P>  <P>
1412  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
1413  defined in the header file:  the subject when subpattern <i>n</i> has not been used at all. For example, if
1414    the string "abc" is matched against the pattern (a|(z))(bc) the return from the
1415    function is 4, and subpatterns 1 and 3 are matched, but 2 is not. When this
1416    happens, both values in the offset pairs corresponding to unused subpatterns
1417    are set to -1.
1418  </P>  </P>
1419  <P>  <P>
1420  <pre>  Offset values that correspond to unused subpatterns at the end of the
1421    PCRE_ERROR_NOMATCH        (-1)  expression are also set to -1. For example, if the string "abc" is matched
1422  </PRE>  against the pattern (abc)(x(yz)?)? subpatterns 2 and 3 are not matched. The
1423    return from the function is 2, because the highest used capturing subpattern
1424    number is 1. However, you can refer to the offsets for the second and third
1425    capturing subpatterns if you wish (assuming the vector is large enough, of
1426    course).
1427  </P>  </P>
1428  <P>  <P>
1429  The subject string did not match the pattern.  Some convenience functions are provided for extracting the captured substrings
1430  </P>  as separate strings. These are described below.
1431    <a name="errorlist"></a></P>
1432    <br><b>
1433    Error return values from <b>pcre_exec()</b>
1434    </b><br>
1435  <P>  <P>
1436    If <b>pcre_exec()</b> fails, it returns a negative number. The following are
1437    defined in the header file:
1438    <pre>
1439      PCRE_ERROR_NOMATCH        (-1)
1440    </pre>
1441    The subject string did not match the pattern.
1442  <pre>  <pre>
1443    PCRE_ERROR_NULL           (-2)    PCRE_ERROR_NULL           (-2)
1444  </PRE>  </pre>
 </P>  
 <P>  
1445  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
1446  NULL and <i>ovecsize</i> was not zero.  NULL and <i>ovecsize</i> was not zero.
 </P>  
 <P>  
1447  <pre>  <pre>
1448    PCRE_ERROR_BADOPTION      (-3)    PCRE_ERROR_BADOPTION      (-3)
1449  </PRE>  </pre>
 </P>  
 <P>  
1450  An unrecognized bit was set in the <i>options</i> argument.  An unrecognized bit was set in the <i>options</i> argument.
 </P>  
 <P>  
1451  <pre>  <pre>
1452    PCRE_ERROR_BADMAGIC       (-4)    PCRE_ERROR_BADMAGIC       (-4)
1453  </PRE>  </pre>
 </P>  
 <P>  
1454  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
1455  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
1456  magic number isn't present.  compiled in an environment of one endianness is run in an environment with the
1457  </P>  other endianness. This is the error that PCRE gives when the magic number is
1458  <P>  not present.
1459  <pre>  <pre>
1460    PCRE_ERROR_UNKNOWN_NODE   (-5)    PCRE_ERROR_UNKNOWN_OPCODE (-5)
1461  </PRE>  </pre>
 </P>  
 <P>  
1462  While running the pattern match, an unknown item was encountered in the  While running the pattern match, an unknown item was encountered in the
1463  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
1464  of the compiled pattern.  of the compiled pattern.
 </P>  
 <P>  
1465  <pre>  <pre>
1466    PCRE_ERROR_NOMEMORY       (-6)    PCRE_ERROR_NOMEMORY       (-6)
1467  </PRE>  </pre>
 </P>  
 <P>  
1468  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
1469  <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
1470  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
1471  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
1472  the end of matching.  automatically freed at the end of matching.
 </P>  
 <P>  
1473  <pre>  <pre>
1474    PCRE_ERROR_NOSUBSTRING    (-7)    PCRE_ERROR_NOSUBSTRING    (-7)
1475  </PRE>  </pre>
 </P>  
 <P>  
1476  This error is used by the <b>pcre_copy_substring()</b>,  This error is used by the <b>pcre_copy_substring()</b>,
1477  <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
1478  below). It is never returned by <b>pcre_exec()</b>.  below). It is never returned by <b>pcre_exec()</b>.
 </P>  
 <P>  
1479  <pre>  <pre>
1480    PCRE_ERROR_MATCHLIMIT     (-8)    PCRE_ERROR_MATCHLIMIT     (-8)
1481  </PRE>  </pre>
1482  </P>  The backtracking limit, as specified by the <i>match_limit</i> field in a
1483  <P>  <b>pcre_extra</b> structure (or defaulted) was reached. See the description
1484  The recursion and backtracking limit, as specified by the <i>match_limit</i>  above.
 field in a <b>pcre_extra</b> structure (or defaulted) was reached. See the  
 description above.  
 </P>  
 <P>  
1485  <pre>  <pre>
1486    PCRE_ERROR_CALLOUT        (-9)    PCRE_ERROR_CALLOUT        (-9)
1487  </PRE>  </pre>
 </P>  
 <P>  
1488  This error is never generated by <b>pcre_exec()</b> itself. It is provided for  This error is never generated by <b>pcre_exec()</b> itself. It is provided for
1489  use by callout functions that want to yield a distinctive error code. See the  use by callout functions that want to yield a distinctive error code. See the
1490  <b>pcrecallout</b> documentation for details.  <a href="pcrecallout.html"><b>pcrecallout</b></a>
1491  </P>  documentation for details.
 <P>  
1492  <pre>  <pre>
1493    PCRE_ERROR_BADUTF8        (-10)    PCRE_ERROR_BADUTF8        (-10)
1494  </PRE>  </pre>
 </P>  
 <P>  
1495  A string that contains an invalid UTF-8 byte sequence was passed as a subject.  A string that contains an invalid UTF-8 byte sequence was passed as a subject.
 </P>  
 <P>  
1496  <pre>  <pre>
1497    PCRE_ERROR_BADUTF8_OFFSET (-11)    PCRE_ERROR_BADUTF8_OFFSET (-11)
1498  </PRE>  </pre>
 </P>  
 <P>  
1499  The UTF-8 byte sequence that was passed as a subject was valid, but the value  The UTF-8 byte sequence that was passed as a subject was valid, but the value
1500  of <i>startoffset</i> did not point to the beginning of a UTF-8 character.  of <i>startoffset</i> did not point to the beginning of a UTF-8 character.
1501    <pre>
1502      PCRE_ERROR_PARTIAL        (-12)
1503    </pre>
1504    The subject string did not match, but it did match partially. See the
1505    <a href="pcrepartial.html"><b>pcrepartial</b></a>
1506    documentation for details of partial matching.
1507    <pre>
1508      PCRE_ERROR_BADPARTIAL     (-13)
1509    </pre>
1510    The PCRE_PARTIAL option was used with a compiled pattern containing items that
1511    are not supported for partial matching. See the
1512    <a href="pcrepartial.html"><b>pcrepartial</b></a>
1513    documentation for details of partial matching.
1514    <pre>
1515      PCRE_ERROR_INTERNAL       (-14)
1516    </pre>
1517    An unexpected internal error has occurred. This error could be caused by a bug
1518    in PCRE or by overwriting of the compiled pattern.
1519    <pre>
1520      PCRE_ERROR_BADCOUNT       (-15)
1521    </pre>
1522    This error is given if the value of the <i>ovecsize</i> argument is negative.
1523    <pre>
1524      PCRE_ERROR_RECURSIONLIMIT (-21)
1525    </pre>
1526    The internal recursion limit, as specified by the <i>match_limit_recursion</i>
1527    field in a <b>pcre_extra</b> structure (or defaulted) was reached. See the
1528    description above.
1529    <pre>
1530      PCRE_ERROR_BADNEWLINE     (-23)
1531    </pre>
1532    An invalid combination of PCRE_NEWLINE_<i>xxx</i> options was given.
1533    </P>
1534    <P>
1535    Error numbers -16 to -20 and -22 are not used by <b>pcre_exec()</b>.
1536  </P>  </P>
1537  <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>
1538  <P>  <P>
1539  <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>
1540  <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 1212  Captured substrings can be accessed dire Line 1556  Captured substrings can be accessed dire
1556  <b>pcre_get_substring_list()</b> are provided for extracting captured substrings  <b>pcre_get_substring_list()</b> are provided for extracting captured substrings
1557  as new, separate, zero-terminated strings. These functions identify substrings  as new, separate, zero-terminated strings. These functions identify substrings
1558  by number. The next section describes functions for extracting named  by number. The next section describes functions for extracting named
1559  substrings. A substring that contains a binary zero is correctly extracted and  substrings.
1560  has a further zero added on the end, but the result is not, of course,  </P>
1561  a C string.  <P>
1562    A substring that contains a binary zero is correctly extracted and has a
1563    further zero added on the end, but the result is not, of course, a C string.
1564    However, you can process such a string by referring to the length that is
1565    returned by <b>pcre_copy_substring()</b> and <b>pcre_get_substring()</b>.
1566    Unfortunately, the interface to <b>pcre_get_substring_list()</b> is not adequate
1567    for handling strings containing binary zeros, because the end of the final
1568    string is not independently indicated.
1569  </P>  </P>
1570  <P>  <P>
1571  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:
1572  <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,
1573  <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
1574  <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
1575  captured by the match, including the substring that matched the entire regular  captured by the match, including the substring that matched the entire regular
1576  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
1577  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
1578  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
1579  the vector divided by three.  number of elements in the vector divided by three.
1580  </P>  </P>
1581  <P>  <P>
1582  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>
1583  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
1584  value of zero extracts the substring that matched the entire pattern, while  value of zero extracts the substring that matched the entire pattern, whereas
1585  higher values extract the captured substrings. For <b>pcre_copy_substring()</b>,  higher values extract the captured substrings. For <b>pcre_copy_substring()</b>,
1586  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
1587  <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
1588  obtained via <b>pcre_malloc</b>, and its address is returned via  obtained via <b>pcre_malloc</b>, and its address is returned via
1589  <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
1590  including the terminating zero, or one of  including the terminating zero, or one of these error codes:
 </P>  
 <P>  
1591  <pre>  <pre>
1592    PCRE_ERROR_NOMEMORY       (-6)    PCRE_ERROR_NOMEMORY       (-6)
1593  </PRE>  </pre>
 </P>  
 <P>  
1594  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
1595  memory failed for <b>pcre_get_substring()</b>.  memory failed for <b>pcre_get_substring()</b>.
 </P>  
 <P>  
1596  <pre>  <pre>
1597    PCRE_ERROR_NOSUBSTRING    (-7)    PCRE_ERROR_NOSUBSTRING    (-7)
1598  </PRE>  </pre>
 </P>  
 <P>  
1599  There is no substring whose number is <i>stringnumber</i>.  There is no substring whose number is <i>stringnumber</i>.
1600  </P>  </P>
1601  <P>  <P>
1602  The <b>pcre_get_substring_list()</b> function extracts all available substrings  The <b>pcre_get_substring_list()</b> function extracts all available substrings
1603  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
1604  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
1605  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
1606  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
1607  function is zero if all went well, or  function is zero if all went well, or the error code
 </P>  
 <P>  
1608  <pre>  <pre>
1609    PCRE_ERROR_NOMEMORY       (-6)    PCRE_ERROR_NOMEMORY       (-6)
1610  </PRE>  </pre>
 </P>  
 <P>  
1611  if the attempt to get the memory block failed.  if the attempt to get the memory block failed.
1612  </P>  </P>
1613  <P>  <P>
# Line 1286  a previous call of <b>pcre_get_substring Line 1625  a previous call of <b>pcre_get_substring
1625  <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
1626  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
1627  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
1628  linked via a special interface to another programming language which cannot use  linked via a special interface to another programming language that cannot use
1629  <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
1630  provided.  provided.
1631  </P>  </P>
1632  <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>
1633    <P>
1634    <b>int pcre_get_stringnumber(const pcre *<i>code</i>,</b>
1635    <b>const char *<i>name</i>);</b>
1636    </P>
1637  <P>  <P>
1638  <b>int pcre_copy_named_substring(const pcre *<i>code</i>,</b>  <b>int pcre_copy_named_substring(const pcre *<i>code</i>,</b>
1639  <b>const char *<i>subject</i>, int *<i>ovector</i>,</b>  <b>const char *<i>subject</i>, int *<i>ovector</i>,</b>
# Line 1298  provided. Line 1641  provided.
1641  <b>char *<i>buffer</i>, int <i>buffersize</i>);</b>  <b>char *<i>buffer</i>, int <i>buffersize</i>);</b>
1642  </P>  </P>
1643  <P>  <P>
 <b>int pcre_get_stringnumber(const pcre *<i>code</i>,</b>  
 <b>const char *<i>name</i>);</b>  
 </P>  
 <P>  
1644  <b>int pcre_get_named_substring(const pcre *<i>code</i>,</b>  <b>int pcre_get_named_substring(const pcre *<i>code</i>,</b>
1645  <b>const char *<i>subject</i>, int *<i>ovector</i>,</b>  <b>const char *<i>subject</i>, int *<i>ovector</i>,</b>
1646  <b>int <i>stringcount</i>, const char *<i>stringname</i>,</b>  <b>int <i>stringcount</i>, const char *<i>stringname</i>,</b>
1647  <b>const char **<i>stringptr</i>);</b>  <b>const char **<i>stringptr</i>);</b>
1648  </P>  </P>
1649  <P>  <P>
1650  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.
1651  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>  
1652  <pre>  <pre>
1653    ab(?&#60;xxx&#62;\d+)...    (a+)b(?&#60;xxx&#62;\d+)...
1654  </PRE>  </pre>
1655  </P>  the number of the subpattern called "xxx" is 2. If the name is known to be
1656  <P>  unique (PCRE_DUPNAMES was not set), you can find the number from the name by
1657  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
1658  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
1659  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
1660  whole job.  that name.
1661  </P>  </P>
1662  <P>  <P>
1663  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
1664  <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
1665  extract by number, and so are not re-described here. There are just two  two functions that do the whole job.
1666  differences.  </P>
1667    <P>
1668    Most of the arguments of <b>pcre_copy_named_substring()</b> and
1669    <b>pcre_get_named_substring()</b> are the same as those for the similarly named
1670    functions that extract by number. As these are described in the previous
1671    section, they are not re-described here. There are just two differences:
1672  </P>  </P>
1673  <P>  <P>
1674  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 1337  translation table. Line 1678  translation table.
1678  </P>  </P>
1679  <P>  <P>
1680  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
1681  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
1682  appropriate.  appropriate. <b>NOTE:</b> If PCRE_DUPNAMES is set and there are duplicate names,
1683    the behaviour may not be what you want (see the next section).
1684    </P>
1685    <br><a name="SEC17" href="#TOC1">DUPLICATE SUBPATTERN NAMES</a><br>
1686    <P>
1687    <b>int pcre_get_stringtable_entries(const pcre *<i>code</i>,</b>
1688    <b>const char *<i>name</i>, char **<i>first</i>, char **<i>last</i>);</b>
1689    </P>
1690    <P>
1691    When a pattern is compiled with the PCRE_DUPNAMES option, names for subpatterns
1692    are not required to be unique. Normally, patterns with duplicate names are such
1693    that in any one match, only one of the named subpatterns participates. An
1694    example is shown in the
1695    <a href="pcrepattern.html"><b>pcrepattern</b></a>
1696    documentation.
1697    </P>
1698    <P>
1699    When duplicates are present, <b>pcre_copy_named_substring()</b> and
1700    <b>pcre_get_named_substring()</b> return the first substring corresponding to
1701    the given name that is set. If none are set, PCRE_ERROR_NOSUBSTRING (-7) is
1702    returned; no data is returned. The <b>pcre_get_stringnumber()</b> function
1703    returns one of the numbers that are associated with the name, but it is not
1704    defined which it is.
1705    </P>
1706    <P>
1707    If you want to get full details of all captured substrings for a given name,
1708    you must use the <b>pcre_get_stringtable_entries()</b> function. The first
1709    argument is the compiled pattern, and the second is the name. The third and
1710    fourth are pointers to variables which are updated by the function. After it
1711    has run, they point to the first and last entries in the name-to-number table
1712    for the given name. The function itself returns the length of each entry, or
1713    PCRE_ERROR_NOSUBSTRING (-7) if there are none. The format of the table is
1714    described above in the section entitled <i>Information about a pattern</i>.
1715    Given all the relevant entries for the name, you can extract each of their
1716    numbers, and hence the captured data, if any.
1717    </P>
1718    <br><a name="SEC18" href="#TOC1">FINDING ALL POSSIBLE MATCHES</a><br>
1719    <P>
1720    The traditional matching function uses a similar algorithm to Perl, which stops
1721    when it finds the first match, starting at a given point in the subject. If you
1722    want to find all possible matches, or the longest possible match, consider
1723    using the alternative matching function (see below) instead. If you cannot use
1724    the alternative function, but still need to find all possible matches, you
1725    can kludge it up by making use of the callout facility, which is described in
1726    the
1727    <a href="pcrecallout.html"><b>pcrecallout</b></a>
1728    documentation.
1729  </P>  </P>
1730  <P>  <P>
1731  Last updated: 09 December 2003  What you have to do is to insert a callout right at the end of the pattern.
1732    When your callout function is called, extract and save the current matched
1733    substring. Then return 1, which forces <b>pcre_exec()</b> to backtrack and try
1734    other alternatives. Ultimately, when it runs out of matches, <b>pcre_exec()</b>
1735    will yield PCRE_ERROR_NOMATCH.
1736    <a name="dfamatch"></a></P>
1737    <br><a name="SEC19" href="#TOC1">MATCHING A PATTERN: THE ALTERNATIVE FUNCTION</a><br>
1738    <P>
1739    <b>int pcre_dfa_exec(const pcre *<i>code</i>, const pcre_extra *<i>extra</i>,</b>
1740    <b>const char *<i>subject</i>, int <i>length</i>, int <i>startoffset</i>,</b>
1741    <b>int <i>options</i>, int *<i>ovector</i>, int <i>ovecsize</i>,</b>
1742    <b>int *<i>workspace</i>, int <i>wscount</i>);</b>
1743    </P>
1744    <P>
1745    The function <b>pcre_dfa_exec()</b> is called to match a subject string against
1746    a compiled pattern, using a matching algorithm that scans the subject string
1747    just once, and does not backtrack. This has different characteristics to the
1748    normal algorithm, and is not compatible with Perl. Some of the features of PCRE
1749    patterns are not supported. Nevertheless, there are times when this kind of
1750    matching can be useful. For a discussion of the two matching algorithms, see
1751    the
1752    <a href="pcrematching.html"><b>pcrematching</b></a>
1753    documentation.
1754    </P>
1755    <P>
1756    The arguments for the <b>pcre_dfa_exec()</b> function are the same as for
1757    <b>pcre_exec()</b>, plus two extras. The <i>ovector</i> argument is used in a
1758    different way, and this is described below. The other common arguments are used
1759    in the same way as for <b>pcre_exec()</b>, so their description is not repeated
1760    here.
1761    </P>
1762    <P>
1763    The two additional arguments provide workspace for the function. The workspace
1764    vector should contain at least 20 elements. It is used for keeping track of
1765    multiple paths through the pattern tree. More workspace will be needed for
1766    patterns and subjects where there are a lot of potential matches.
1767    </P>
1768    <P>
1769    Here is an example of a simple call to <b>pcre_dfa_exec()</b>:
1770    <pre>
1771      int rc;
1772      int ovector[10];
1773      int wspace[20];
1774      rc = pcre_dfa_exec(
1775        re,             /* result of pcre_compile() */
1776        NULL,           /* we didn't study the pattern */
1777        "some string",  /* the subject string */
1778        11,             /* the length of the subject string */
1779        0,              /* start at offset 0 in the subject */
1780        0,              /* default options */
1781        ovector,        /* vector of integers for substring information */
1782        10,             /* number of elements (NOT size in bytes) */
1783        wspace,         /* working space vector */
1784        20);            /* number of elements (NOT size in bytes) */
1785    </PRE>
1786    </P>
1787    <br><b>
1788    Option bits for <b>pcre_dfa_exec()</b>
1789    </b><br>
1790    <P>
1791    The unused bits of the <i>options</i> argument for <b>pcre_dfa_exec()</b> must be
1792    zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_<i>xxx</i>,
1793    PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NO_UTF8_CHECK, PCRE_PARTIAL,
1794    PCRE_DFA_SHORTEST, and PCRE_DFA_RESTART. All but the last three of these are
1795    the same as for <b>pcre_exec()</b>, so their description is not repeated here.
1796    <pre>
1797      PCRE_PARTIAL
1798    </pre>
1799    This has the same general effect as it does for <b>pcre_exec()</b>, but the
1800    details are slightly different. When PCRE_PARTIAL is set for
1801    <b>pcre_dfa_exec()</b>, the return code PCRE_ERROR_NOMATCH is converted into
1802    PCRE_ERROR_PARTIAL if the end of the subject is reached, there have been no
1803    complete matches, but there is still at least one matching possibility. The
1804    portion of the string that provided the partial match is set as the first
1805    matching string.
1806    <pre>
1807      PCRE_DFA_SHORTEST
1808    </pre>
1809    Setting the PCRE_DFA_SHORTEST option causes the matching algorithm to stop as
1810    soon as it has found one match. Because of the way the alternative algorithm
1811    works, this is necessarily the shortest possible match at the first possible
1812    matching point in the subject string.
1813    <pre>
1814      PCRE_DFA_RESTART
1815    </pre>
1816    When <b>pcre_dfa_exec()</b> is called with the PCRE_PARTIAL option, and returns
1817    a partial match, it is possible to call it again, with additional subject
1818    characters, and have it continue with the same match. The PCRE_DFA_RESTART
1819    option requests this action; when it is set, the <i>workspace</i> and
1820    <i>wscount</i> options must reference the same vector as before because data
1821    about the match so far is left in them after a partial match. There is more
1822    discussion of this facility in the
1823    <a href="pcrepartial.html"><b>pcrepartial</b></a>
1824    documentation.
1825    </P>
1826    <br><b>
1827    Successful returns from <b>pcre_dfa_exec()</b>
1828    </b><br>
1829    <P>
1830    When <b>pcre_dfa_exec()</b> succeeds, it may have matched more than one
1831    substring in the subject. Note, however, that all the matches from one run of
1832    the function start at the same point in the subject. The shorter matches are
1833    all initial substrings of the longer matches. For example, if the pattern
1834    <pre>
1835      &#60;.*&#62;
1836    </pre>
1837    is matched against the string
1838    <pre>
1839      This is &#60;something&#62; &#60;something else&#62; &#60;something further&#62; no more
1840    </pre>
1841    the three matched strings are
1842    <pre>
1843      &#60;something&#62;
1844      &#60;something&#62; &#60;something else&#62;
1845      &#60;something&#62; &#60;something else&#62; &#60;something further&#62;
1846    </pre>
1847    On success, the yield of the function is a number greater than zero, which is
1848    the number of matched substrings. The substrings themselves are returned in
1849    <i>ovector</i>. Each string uses two elements; the first is the offset to the
1850    start, and the second is the offset to the end. In fact, all the strings have
1851    the same start offset. (Space could have been saved by giving this only once,
1852    but it was decided to retain some compatibility with the way <b>pcre_exec()</b>
1853    returns data, even though the meaning of the strings is different.)
1854    </P>
1855    <P>
1856    The strings are returned in reverse order of length; that is, the longest
1857    matching string is given first. If there were too many matches to fit into
1858    <i>ovector</i>, the yield of the function is zero, and the vector is filled with
1859    the longest matches.
1860    </P>
1861    <br><b>
1862    Error returns from <b>pcre_dfa_exec()</b>
1863    </b><br>
1864    <P>
1865    The <b>pcre_dfa_exec()</b> function returns a negative number when it fails.
1866    Many of the errors are the same as for <b>pcre_exec()</b>, and these are
1867    described
1868    <a href="#errorlist">above.</a>
1869    There are in addition the following errors that are specific to
1870    <b>pcre_dfa_exec()</b>:
1871    <pre>
1872      PCRE_ERROR_DFA_UITEM      (-16)
1873    </pre>
1874    This return is given if <b>pcre_dfa_exec()</b> encounters an item in the pattern
1875    that it does not support, for instance, the use of \C or a back reference.
1876    <pre>
1877      PCRE_ERROR_DFA_UCOND      (-17)
1878    </pre>
1879    This return is given if <b>pcre_dfa_exec()</b> encounters a condition item that
1880    uses a back reference for the condition, or a test for recursion in a specific
1881    group. These are not supported.
1882    <pre>
1883      PCRE_ERROR_DFA_UMLIMIT    (-18)
1884    </pre>
1885    This return is given if <b>pcre_dfa_exec()</b> is called with an <i>extra</i>
1886    block that contains a setting of the <i>match_limit</i> field. This is not
1887    supported (it is meaningless).
1888    <pre>
1889      PCRE_ERROR_DFA_WSSIZE     (-19)
1890    </pre>
1891    This return is given if <b>pcre_dfa_exec()</b> runs out of space in the
1892    <i>workspace</i> vector.
1893    <pre>
1894      PCRE_ERROR_DFA_RECURSE    (-20)
1895    </pre>
1896    When a recursive subpattern is processed, the matching function calls itself
1897    recursively, using private vectors for <i>ovector</i> and <i>workspace</i>. This
1898    error is given if the output vector is not large enough. This should be
1899    extremely rare, as a vector of size 1000 is used.
1900    </P>
1901    <br><a name="SEC20" href="#TOC1">SEE ALSO</a><br>
1902    <P>
1903    <b>pcrebuild</b>(3), <b>pcrecallout</b>(3), <b>pcrecpp(3)</b>(3),
1904    <b>pcrematching</b>(3), <b>pcrepartial</b>(3), <b>pcreposix</b>(3),
1905    <b>pcreprecompile</b>(3), <b>pcresample</b>(3), <b>pcrestack</b>(3).
1906    </P>
1907    <br><a name="SEC21" href="#TOC1">AUTHOR</a><br>
1908    <P>
1909    Philip Hazel
1910    <br>
1911    University Computing Service
1912    <br>
1913    Cambridge CB2 3QH, England.
1914    <br>
1915    </P>
1916    <br><a name="SEC22" href="#TOC1">REVISION</a><br>
1917    <P>
1918    Last updated: 21 August 2007
1919    <br>
1920    Copyright &copy; 1997-2007 University of Cambridge.
1921  <br>  <br>
1922  Copyright &copy; 1997-2003 University of Cambridge.  <p>
1923    Return to the <a href="index.html">PCRE index page</a>.
1924    </p>

Legend:
Removed from v.73  
changed lines
  Added in v.227

  ViewVC Help
Powered by ViewVC 1.1.5