Annotation of embedaddon/pcre/doc/pcreapi.3, revision 1.1.1.3
1.1.1.3 ! misho 1: .TH PCREAPI 3 "04 May 2012" "PCRE 8.31"
1.1 misho 2: .SH NAME
3: PCRE - Perl-compatible regular expressions
1.1.1.2 misho 4: .sp
5: .B #include <pcre.h>
6: .
7: .
1.1 misho 8: .SH "PCRE NATIVE API BASIC FUNCTIONS"
9: .rs
10: .sp
11: .SM
12: .B pcre *pcre_compile(const char *\fIpattern\fP, int \fIoptions\fP,
13: .ti +5n
14: .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
15: .ti +5n
16: .B const unsigned char *\fItableptr\fP);
17: .PP
18: .B pcre *pcre_compile2(const char *\fIpattern\fP, int \fIoptions\fP,
19: .ti +5n
20: .B int *\fIerrorcodeptr\fP,
21: .ti +5n
22: .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
23: .ti +5n
24: .B const unsigned char *\fItableptr\fP);
25: .PP
26: .B pcre_extra *pcre_study(const pcre *\fIcode\fP, int \fIoptions\fP,
27: .ti +5n
28: .B const char **\fIerrptr\fP);
29: .PP
30: .B void pcre_free_study(pcre_extra *\fIextra\fP);
31: .PP
32: .B int pcre_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
33: .ti +5n
34: .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
35: .ti +5n
36: .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP);
37: .PP
38: .B int pcre_dfa_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
39: .ti +5n
40: .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
41: .ti +5n
42: .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP,
43: .ti +5n
44: .B int *\fIworkspace\fP, int \fIwscount\fP);
1.1.1.2 misho 45: .
46: .
47: .SH "PCRE NATIVE API STRING EXTRACTION FUNCTIONS"
48: .rs
49: .sp
1.1 misho 50: .B int pcre_copy_named_substring(const pcre *\fIcode\fP,
51: .ti +5n
52: .B const char *\fIsubject\fP, int *\fIovector\fP,
53: .ti +5n
54: .B int \fIstringcount\fP, const char *\fIstringname\fP,
55: .ti +5n
56: .B char *\fIbuffer\fP, int \fIbuffersize\fP);
57: .PP
58: .B int pcre_copy_substring(const char *\fIsubject\fP, int *\fIovector\fP,
59: .ti +5n
60: .B int \fIstringcount\fP, int \fIstringnumber\fP, char *\fIbuffer\fP,
61: .ti +5n
62: .B int \fIbuffersize\fP);
63: .PP
64: .B int pcre_get_named_substring(const pcre *\fIcode\fP,
65: .ti +5n
66: .B const char *\fIsubject\fP, int *\fIovector\fP,
67: .ti +5n
68: .B int \fIstringcount\fP, const char *\fIstringname\fP,
69: .ti +5n
70: .B const char **\fIstringptr\fP);
71: .PP
72: .B int pcre_get_stringnumber(const pcre *\fIcode\fP,
73: .ti +5n
74: .B const char *\fIname\fP);
75: .PP
76: .B int pcre_get_stringtable_entries(const pcre *\fIcode\fP,
77: .ti +5n
78: .B const char *\fIname\fP, char **\fIfirst\fP, char **\fIlast\fP);
79: .PP
80: .B int pcre_get_substring(const char *\fIsubject\fP, int *\fIovector\fP,
81: .ti +5n
82: .B int \fIstringcount\fP, int \fIstringnumber\fP,
83: .ti +5n
84: .B const char **\fIstringptr\fP);
85: .PP
86: .B int pcre_get_substring_list(const char *\fIsubject\fP,
87: .ti +5n
88: .B int *\fIovector\fP, int \fIstringcount\fP, "const char ***\fIlistptr\fP);"
89: .PP
90: .B void pcre_free_substring(const char *\fIstringptr\fP);
91: .PP
92: .B void pcre_free_substring_list(const char **\fIstringptr\fP);
1.1.1.2 misho 93: .
94: .
95: .SH "PCRE NATIVE API AUXILIARY FUNCTIONS"
96: .rs
97: .sp
98: .B pcre_jit_stack *pcre_jit_stack_alloc(int \fIstartsize\fP, int \fImaxsize\fP);
99: .PP
100: .B void pcre_jit_stack_free(pcre_jit_stack *\fIstack\fP);
101: .PP
102: .B void pcre_assign_jit_stack(pcre_extra *\fIextra\fP,
103: .ti +5n
104: .B pcre_jit_callback \fIcallback\fP, void *\fIdata\fP);
1.1 misho 105: .PP
106: .B const unsigned char *pcre_maketables(void);
107: .PP
108: .B int pcre_fullinfo(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
109: .ti +5n
110: .B int \fIwhat\fP, void *\fIwhere\fP);
111: .PP
112: .B int pcre_refcount(pcre *\fIcode\fP, int \fIadjust\fP);
113: .PP
114: .B int pcre_config(int \fIwhat\fP, void *\fIwhere\fP);
115: .PP
1.1.1.2 misho 116: .B const char *pcre_version(void);
117: .PP
118: .B int pcre_pattern_to_host_byte_order(pcre *\fIcode\fP,
119: .ti +5n
120: .B pcre_extra *\fIextra\fP, const unsigned char *\fItables\fP);
1.1 misho 121: .
122: .
123: .SH "PCRE NATIVE API INDIRECTED FUNCTIONS"
124: .rs
125: .sp
126: .B void *(*pcre_malloc)(size_t);
127: .PP
128: .B void (*pcre_free)(void *);
129: .PP
130: .B void *(*pcre_stack_malloc)(size_t);
131: .PP
132: .B void (*pcre_stack_free)(void *);
133: .PP
134: .B int (*pcre_callout)(pcre_callout_block *);
135: .
136: .
1.1.1.2 misho 137: .SH "PCRE 8-BIT AND 16-BIT LIBRARIES"
138: .rs
139: .sp
140: From release 8.30, PCRE can be compiled as a library for handling 16-bit
141: character strings as well as, or instead of, the original library that handles
142: 8-bit character strings. To avoid too much complication, this document
143: describes the 8-bit versions of the functions, with only occasional references
144: to the 16-bit library.
145: .P
146: The 16-bit functions operate in the same way as their 8-bit counterparts; they
147: just use different data types for their arguments and results, and their names
148: start with \fBpcre16_\fP instead of \fBpcre_\fP. For every option that has UTF8
149: in its name (for example, PCRE_UTF8), there is a corresponding 16-bit name with
150: UTF8 replaced by UTF16. This facility is in fact just cosmetic; the 16-bit
151: option names define the same bit values.
152: .P
153: References to bytes and UTF-8 in this document should be read as references to
154: 16-bit data quantities and UTF-16 when using the 16-bit library, unless
155: specified otherwise. More details of the specific differences for the 16-bit
156: library are given in the
157: .\" HREF
158: \fBpcre16\fP
159: .\"
160: page.
161: .
162: .
1.1 misho 163: .SH "PCRE API OVERVIEW"
164: .rs
165: .sp
166: PCRE has its own native API, which is described in this document. There are
1.1.1.2 misho 167: also some wrapper functions (for the 8-bit library only) that correspond to the
168: POSIX regular expression API, but they do not give access to all the
169: functionality. They are described in the
1.1 misho 170: .\" HREF
171: \fBpcreposix\fP
172: .\"
173: documentation. Both of these APIs define a set of C function calls. A C++
1.1.1.2 misho 174: wrapper (again for the 8-bit library only) is also distributed with PCRE. It is
175: documented in the
1.1 misho 176: .\" HREF
177: \fBpcrecpp\fP
178: .\"
179: page.
180: .P
181: The native API C function prototypes are defined in the header file
1.1.1.2 misho 182: \fBpcre.h\fP, and on Unix-like systems the (8-bit) library itself is called
183: \fBlibpcre\fP. It can normally be accessed by adding \fB-lpcre\fP to the
184: command for linking an application that uses PCRE. The header file defines the
185: macros PCRE_MAJOR and PCRE_MINOR to contain the major and minor release numbers
186: for the library. Applications can use these to include support for different
187: releases of PCRE.
1.1 misho 188: .P
189: In a Windows environment, if you want to statically link an application program
190: against a non-dll \fBpcre.a\fP file, you must define PCRE_STATIC before
191: including \fBpcre.h\fP or \fBpcrecpp.h\fP, because otherwise the
192: \fBpcre_malloc()\fP and \fBpcre_free()\fP exported functions will be declared
193: \fB__declspec(dllimport)\fP, with unwanted results.
194: .P
195: The functions \fBpcre_compile()\fP, \fBpcre_compile2()\fP, \fBpcre_study()\fP,
196: and \fBpcre_exec()\fP are used for compiling and matching regular expressions
197: in a Perl-compatible manner. A sample program that demonstrates the simplest
198: way of using them is provided in the file called \fIpcredemo.c\fP in the PCRE
199: source distribution. A listing of this program is given in the
200: .\" HREF
201: \fBpcredemo\fP
202: .\"
203: documentation, and the
204: .\" HREF
205: \fBpcresample\fP
206: .\"
207: documentation describes how to compile and run it.
208: .P
209: Just-in-time compiler support is an optional feature of PCRE that can be built
210: in appropriate hardware environments. It greatly speeds up the matching
211: performance of many patterns. Simple programs can easily request that it be
212: used if available, by setting an option that is ignored when it is not
213: relevant. More complicated programs might need to make use of the functions
214: \fBpcre_jit_stack_alloc()\fP, \fBpcre_jit_stack_free()\fP, and
215: \fBpcre_assign_jit_stack()\fP in order to control the JIT code's memory usage.
216: These functions are discussed in the
217: .\" HREF
218: \fBpcrejit\fP
219: .\"
220: documentation.
221: .P
222: A second matching function, \fBpcre_dfa_exec()\fP, which is not
223: Perl-compatible, is also provided. This uses a different algorithm for the
224: matching. The alternative algorithm finds all possible matches (at a given
225: point in the subject), and scans the subject just once (unless there are
226: lookbehind assertions). However, this algorithm does not return captured
227: substrings. A description of the two matching algorithms and their advantages
228: and disadvantages is given in the
229: .\" HREF
230: \fBpcrematching\fP
231: .\"
232: documentation.
233: .P
234: In addition to the main compiling and matching functions, there are convenience
235: functions for extracting captured substrings from a subject string that is
236: matched by \fBpcre_exec()\fP. They are:
237: .sp
238: \fBpcre_copy_substring()\fP
239: \fBpcre_copy_named_substring()\fP
240: \fBpcre_get_substring()\fP
241: \fBpcre_get_named_substring()\fP
242: \fBpcre_get_substring_list()\fP
243: \fBpcre_get_stringnumber()\fP
244: \fBpcre_get_stringtable_entries()\fP
245: .sp
246: \fBpcre_free_substring()\fP and \fBpcre_free_substring_list()\fP are also
247: provided, to free the memory used for extracted strings.
248: .P
249: The function \fBpcre_maketables()\fP is used to build a set of character tables
250: in the current locale for passing to \fBpcre_compile()\fP, \fBpcre_exec()\fP,
251: or \fBpcre_dfa_exec()\fP. This is an optional facility that is provided for
252: specialist use. Most commonly, no special tables are passed, in which case
253: internal tables that are generated when PCRE is built are used.
254: .P
255: The function \fBpcre_fullinfo()\fP is used to find out information about a
1.1.1.2 misho 256: compiled pattern. The function \fBpcre_version()\fP returns a pointer to a
257: string containing the version of PCRE and its date of release.
1.1 misho 258: .P
259: The function \fBpcre_refcount()\fP maintains a reference count in a data block
260: containing a compiled pattern. This is provided for the benefit of
261: object-oriented applications.
262: .P
263: The global variables \fBpcre_malloc\fP and \fBpcre_free\fP initially contain
264: the entry points of the standard \fBmalloc()\fP and \fBfree()\fP functions,
265: respectively. PCRE calls the memory management functions via these variables,
266: so a calling program can replace them if it wishes to intercept the calls. This
267: should be done before calling any PCRE functions.
268: .P
269: The global variables \fBpcre_stack_malloc\fP and \fBpcre_stack_free\fP are also
270: indirections to memory management functions. These special functions are used
271: only when PCRE is compiled to use the heap for remembering data, instead of
272: recursive function calls, when running the \fBpcre_exec()\fP function. See the
273: .\" HREF
274: \fBpcrebuild\fP
275: .\"
276: documentation for details of how to do this. It is a non-standard way of
277: building PCRE, for use in environments that have limited stacks. Because of the
278: greater use of memory management, it runs more slowly. Separate functions are
279: provided so that special-purpose external code can be used for this case. When
280: used, these functions are always called in a stack-like manner (last obtained,
281: first freed), and always for memory blocks of the same size. There is a
282: discussion about PCRE's stack usage in the
283: .\" HREF
284: \fBpcrestack\fP
285: .\"
286: documentation.
287: .P
288: The global variable \fBpcre_callout\fP initially contains NULL. It can be set
289: by the caller to a "callout" function, which PCRE will then call at specified
290: points during a matching operation. Details are given in the
291: .\" HREF
292: \fBpcrecallout\fP
293: .\"
294: documentation.
295: .
296: .
297: .\" HTML <a name="newlines"></a>
298: .SH NEWLINES
299: .rs
300: .sp
301: PCRE supports five different conventions for indicating line breaks in
302: strings: a single CR (carriage return) character, a single LF (linefeed)
303: character, the two-character sequence CRLF, any of the three preceding, or any
304: Unicode newline sequence. The Unicode newline sequences are the three just
1.1.1.3 ! misho 305: mentioned, plus the single characters VT (vertical tab, U+000B), FF (form feed,
1.1 misho 306: U+000C), NEL (next line, U+0085), LS (line separator, U+2028), and PS
307: (paragraph separator, U+2029).
308: .P
309: Each of the first three conventions is used by at least one operating system as
310: its standard newline sequence. When PCRE is built, a default can be specified.
311: The default default is LF, which is the Unix standard. When PCRE is run, the
312: default can be overridden, either when a pattern is compiled, or when it is
313: matched.
314: .P
315: At compile time, the newline convention can be specified by the \fIoptions\fP
316: argument of \fBpcre_compile()\fP, or it can be specified by special text at the
317: start of the pattern itself; this overrides any other settings. See the
318: .\" HREF
319: \fBpcrepattern\fP
320: .\"
321: page for details of the special character sequences.
322: .P
323: In the PCRE documentation the word "newline" is used to mean "the character or
324: pair of characters that indicate a line break". The choice of newline
325: convention affects the handling of the dot, circumflex, and dollar
326: metacharacters, the handling of #-comments in /x mode, and, when CRLF is a
327: recognized line ending sequence, the match position advancement for a
328: non-anchored pattern. There is more detail about this in the
329: .\" HTML <a href="#execoptions">
330: .\" </a>
331: section on \fBpcre_exec()\fP options
332: .\"
333: below.
334: .P
335: The choice of newline convention does not affect the interpretation of
336: the \en or \er escape sequences, nor does it affect what \eR matches, which is
337: controlled in a similar way, but by separate options.
338: .
339: .
340: .SH MULTITHREADING
341: .rs
342: .sp
343: The PCRE functions can be used in multi-threading applications, with the
344: proviso that the memory management functions pointed to by \fBpcre_malloc\fP,
345: \fBpcre_free\fP, \fBpcre_stack_malloc\fP, and \fBpcre_stack_free\fP, and the
346: callout function pointed to by \fBpcre_callout\fP, are shared by all threads.
347: .P
348: The compiled form of a regular expression is not altered during matching, so
349: the same compiled pattern can safely be used by several threads at once.
350: .P
351: If the just-in-time optimization feature is being used, it needs separate
352: memory stack areas for each thread. See the
353: .\" HREF
354: \fBpcrejit\fP
355: .\"
356: documentation for more details.
357: .
358: .
359: .SH "SAVING PRECOMPILED PATTERNS FOR LATER USE"
360: .rs
361: .sp
362: The compiled form of a regular expression can be saved and re-used at a later
363: time, possibly by a different program, and even on a host other than the one on
364: which it was compiled. Details are given in the
365: .\" HREF
366: \fBpcreprecompile\fP
367: .\"
1.1.1.2 misho 368: documentation, which includes a description of the
369: \fBpcre_pattern_to_host_byte_order()\fP function. However, compiling a regular
370: expression with one version of PCRE for use with a different version is not
371: guaranteed to work and may cause crashes.
1.1 misho 372: .
373: .
374: .SH "CHECKING BUILD-TIME OPTIONS"
375: .rs
376: .sp
377: .B int pcre_config(int \fIwhat\fP, void *\fIwhere\fP);
378: .PP
379: The function \fBpcre_config()\fP makes it possible for a PCRE client to
380: discover which optional features have been compiled into the PCRE library. The
381: .\" HREF
382: \fBpcrebuild\fP
383: .\"
384: documentation has more details about these optional features.
385: .P
386: The first argument for \fBpcre_config()\fP is an integer, specifying which
387: information is required; the second argument is a pointer to a variable into
1.1.1.2 misho 388: which the information is placed. The returned value is zero on success, or the
389: negative error code PCRE_ERROR_BADOPTION if the value in the first argument is
390: not recognized. The following information is available:
1.1 misho 391: .sp
392: PCRE_CONFIG_UTF8
393: .sp
394: The output is an integer that is set to one if UTF-8 support is available;
1.1.1.2 misho 395: otherwise it is set to zero. If this option is given to the 16-bit version of
396: this function, \fBpcre16_config()\fP, the result is PCRE_ERROR_BADOPTION.
397: .sp
398: PCRE_CONFIG_UTF16
399: .sp
400: The output is an integer that is set to one if UTF-16 support is available;
401: otherwise it is set to zero. This value should normally be given to the 16-bit
402: version of this function, \fBpcre16_config()\fP. If it is given to the 8-bit
403: version of this function, the result is PCRE_ERROR_BADOPTION.
1.1 misho 404: .sp
405: PCRE_CONFIG_UNICODE_PROPERTIES
406: .sp
407: The output is an integer that is set to one if support for Unicode character
408: properties is available; otherwise it is set to zero.
409: .sp
410: PCRE_CONFIG_JIT
411: .sp
412: The output is an integer that is set to one if support for just-in-time
413: compiling is available; otherwise it is set to zero.
414: .sp
1.1.1.2 misho 415: PCRE_CONFIG_JITTARGET
416: .sp
417: The output is a pointer to a zero-terminated "const char *" string. If JIT
418: support is available, the string contains the name of the architecture for
419: which the JIT compiler is configured, for example "x86 32bit (little endian +
420: unaligned)". If JIT support is not available, the result is NULL.
421: .sp
1.1 misho 422: PCRE_CONFIG_NEWLINE
423: .sp
424: The output is an integer whose value specifies the default character sequence
425: that is recognized as meaning "newline". The four values that are supported
426: are: 10 for LF, 13 for CR, 3338 for CRLF, -2 for ANYCRLF, and -1 for ANY.
427: Though they are derived from ASCII, the same values are returned in EBCDIC
428: environments. The default should normally correspond to the standard sequence
429: for your operating system.
430: .sp
431: PCRE_CONFIG_BSR
432: .sp
433: The output is an integer whose value indicates what character sequences the \eR
434: escape sequence matches by default. A value of 0 means that \eR matches any
435: Unicode line ending sequence; a value of 1 means that \eR matches only CR, LF,
436: or CRLF. The default can be overridden when a pattern is compiled or matched.
437: .sp
438: PCRE_CONFIG_LINK_SIZE
439: .sp
440: The output is an integer that contains the number of bytes used for internal
1.1.1.2 misho 441: linkage in compiled regular expressions. For the 8-bit library, the value can
442: be 2, 3, or 4. For the 16-bit library, the value is either 2 or 4 and is still
443: a number of bytes. The default value of 2 is sufficient for all but the most
444: massive patterns, since it allows the compiled pattern to be up to 64K in size.
445: Larger values allow larger regular expressions to be compiled, at the expense
446: of slower matching.
1.1 misho 447: .sp
448: PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
449: .sp
450: The output is an integer that contains the threshold above which the POSIX
451: interface uses \fBmalloc()\fP for output vectors. Further details are given in
452: the
453: .\" HREF
454: \fBpcreposix\fP
455: .\"
456: documentation.
457: .sp
458: PCRE_CONFIG_MATCH_LIMIT
459: .sp
460: The output is a long integer that gives the default limit for the number of
461: internal matching function calls in a \fBpcre_exec()\fP execution. Further
462: details are given with \fBpcre_exec()\fP below.
463: .sp
464: PCRE_CONFIG_MATCH_LIMIT_RECURSION
465: .sp
466: The output is a long integer that gives the default limit for the depth of
467: recursion when calling the internal matching function in a \fBpcre_exec()\fP
468: execution. Further details are given with \fBpcre_exec()\fP below.
469: .sp
470: PCRE_CONFIG_STACKRECURSE
471: .sp
472: The output is an integer that is set to one if internal recursion when running
473: \fBpcre_exec()\fP is implemented by recursive function calls that use the stack
474: to remember their state. This is the usual way that PCRE is compiled. The
475: output is zero if PCRE was compiled to use blocks of data on the heap instead
476: of recursive function calls. In this case, \fBpcre_stack_malloc\fP and
477: \fBpcre_stack_free\fP are called to manage memory blocks on the heap, thus
478: avoiding the use of the stack.
479: .
480: .
481: .SH "COMPILING A PATTERN"
482: .rs
483: .sp
484: .B pcre *pcre_compile(const char *\fIpattern\fP, int \fIoptions\fP,
485: .ti +5n
486: .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
487: .ti +5n
488: .B const unsigned char *\fItableptr\fP);
489: .sp
490: .B pcre *pcre_compile2(const char *\fIpattern\fP, int \fIoptions\fP,
491: .ti +5n
492: .B int *\fIerrorcodeptr\fP,
493: .ti +5n
494: .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
495: .ti +5n
496: .B const unsigned char *\fItableptr\fP);
497: .P
498: Either of the functions \fBpcre_compile()\fP or \fBpcre_compile2()\fP can be
499: called to compile a pattern into an internal form. The only difference between
500: the two interfaces is that \fBpcre_compile2()\fP has an additional argument,
501: \fIerrorcodeptr\fP, via which a numerical error code can be returned. To avoid
502: too much repetition, we refer just to \fBpcre_compile()\fP below, but the
503: information applies equally to \fBpcre_compile2()\fP.
504: .P
505: The pattern is a C string terminated by a binary zero, and is passed in the
506: \fIpattern\fP argument. A pointer to a single block of memory that is obtained
507: via \fBpcre_malloc\fP is returned. This contains the compiled code and related
508: data. The \fBpcre\fP type is defined for the returned block; this is a typedef
509: for a structure whose contents are not externally defined. It is up to the
510: caller to free the memory (via \fBpcre_free\fP) when it is no longer required.
511: .P
512: Although the compiled code of a PCRE regex is relocatable, that is, it does not
513: depend on memory location, the complete \fBpcre\fP data block is not
514: fully relocatable, because it may contain a copy of the \fItableptr\fP
515: argument, which is an address (see below).
516: .P
517: The \fIoptions\fP argument contains various bit settings that affect the
518: compilation. It should be zero if no options are required. The available
519: options are described below. Some of them (in particular, those that are
520: compatible with Perl, but some others as well) can also be set and unset from
521: within the pattern (see the detailed description in the
522: .\" HREF
523: \fBpcrepattern\fP
524: .\"
525: documentation). For those options that can be different in different parts of
526: the pattern, the contents of the \fIoptions\fP argument specifies their
527: settings at the start of compilation and execution. The PCRE_ANCHORED,
528: PCRE_BSR_\fIxxx\fP, PCRE_NEWLINE_\fIxxx\fP, PCRE_NO_UTF8_CHECK, and
1.1.1.3 ! misho 529: PCRE_NO_START_OPTIMIZE options can be set at the time of matching as well as at
1.1 misho 530: compile time.
531: .P
532: If \fIerrptr\fP is NULL, \fBpcre_compile()\fP returns NULL immediately.
533: Otherwise, if compilation of a pattern fails, \fBpcre_compile()\fP returns
534: NULL, and sets the variable pointed to by \fIerrptr\fP to point to a textual
535: error message. This is a static string that is part of the library. You must
536: not try to free it. Normally, the offset from the start of the pattern to the
537: byte that was being processed when the error was discovered is placed in the
538: variable pointed to by \fIerroffset\fP, which must not be NULL (if it is, an
539: immediate error is given). However, for an invalid UTF-8 string, the offset is
1.1.1.2 misho 540: that of the first byte of the failing character.
1.1 misho 541: .P
1.1.1.2 misho 542: Some errors are not detected until the whole pattern has been scanned; in these
543: cases, the offset passed back is the length of the pattern. Note that the
544: offset is in bytes, not characters, even in UTF-8 mode. It may sometimes point
545: into the middle of a UTF-8 character.
1.1 misho 546: .P
547: If \fBpcre_compile2()\fP is used instead of \fBpcre_compile()\fP, and the
548: \fIerrorcodeptr\fP argument is not NULL, a non-zero error code number is
549: returned via this argument in the event of an error. This is in addition to the
550: textual error message. Error codes and messages are listed below.
551: .P
552: If the final argument, \fItableptr\fP, is NULL, PCRE uses a default set of
553: character tables that are built when PCRE is compiled, using the default C
554: locale. Otherwise, \fItableptr\fP must be an address that is the result of a
555: call to \fBpcre_maketables()\fP. This value is stored with the compiled
556: pattern, and used again by \fBpcre_exec()\fP, unless another table pointer is
557: passed to it. For more discussion, see the section on locale support below.
558: .P
559: This code fragment shows a typical straightforward call to \fBpcre_compile()\fP:
560: .sp
561: pcre *re;
562: const char *error;
563: int erroffset;
564: re = pcre_compile(
565: "^A.*Z", /* the pattern */
566: 0, /* default options */
567: &error, /* for error message */
568: &erroffset, /* for error offset */
569: NULL); /* use default character tables */
570: .sp
571: The following names for option bits are defined in the \fBpcre.h\fP header
572: file:
573: .sp
574: PCRE_ANCHORED
575: .sp
576: If this bit is set, the pattern is forced to be "anchored", that is, it is
577: constrained to match only at the first matching point in the string that is
578: being searched (the "subject string"). This effect can also be achieved by
579: appropriate constructs in the pattern itself, which is the only way to do it in
580: Perl.
581: .sp
582: PCRE_AUTO_CALLOUT
583: .sp
584: If this bit is set, \fBpcre_compile()\fP automatically inserts callout items,
585: all with number 255, before each pattern item. For discussion of the callout
586: facility, see the
587: .\" HREF
588: \fBpcrecallout\fP
589: .\"
590: documentation.
591: .sp
592: PCRE_BSR_ANYCRLF
593: PCRE_BSR_UNICODE
594: .sp
595: These options (which are mutually exclusive) control what the \eR escape
596: sequence matches. The choice is either to match only CR, LF, or CRLF, or to
597: match any Unicode newline sequence. The default is specified when PCRE is
598: built. It can be overridden from within the pattern, or by setting an option
599: when a compiled pattern is matched.
600: .sp
601: PCRE_CASELESS
602: .sp
603: If this bit is set, letters in the pattern match both upper and lower case
604: letters. It is equivalent to Perl's /i option, and it can be changed within a
605: pattern by a (?i) option setting. In UTF-8 mode, PCRE always understands the
606: concept of case for characters whose values are less than 128, so caseless
607: matching is always possible. For characters with higher values, the concept of
608: case is supported if PCRE is compiled with Unicode property support, but not
609: otherwise. If you want to use caseless matching for characters 128 and above,
610: you must ensure that PCRE is compiled with Unicode property support as well as
611: with UTF-8 support.
612: .sp
613: PCRE_DOLLAR_ENDONLY
614: .sp
615: If this bit is set, a dollar metacharacter in the pattern matches only at the
616: end of the subject string. Without this option, a dollar also matches
617: immediately before a newline at the end of the string (but not before any other
618: newlines). The PCRE_DOLLAR_ENDONLY option is ignored if PCRE_MULTILINE is set.
619: There is no equivalent to this option in Perl, and no way to set it within a
620: pattern.
621: .sp
622: PCRE_DOTALL
623: .sp
624: If this bit is set, a dot metacharacter in the pattern matches a character of
625: any value, including one that indicates a newline. However, it only ever
626: matches one character, even if newlines are coded as CRLF. Without this option,
627: a dot does not match when the current position is at a newline. This option is
628: equivalent to Perl's /s option, and it can be changed within a pattern by a
629: (?s) option setting. A negative class such as [^a] always matches newline
630: characters, independent of the setting of this option.
631: .sp
632: PCRE_DUPNAMES
633: .sp
634: If this bit is set, names used to identify capturing subpatterns need not be
635: unique. This can be helpful for certain types of pattern when it is known that
636: only one instance of the named subpattern can ever be matched. There are more
637: details of named subpatterns below; see also the
638: .\" HREF
639: \fBpcrepattern\fP
640: .\"
641: documentation.
642: .sp
643: PCRE_EXTENDED
644: .sp
1.1.1.3 ! misho 645: If this bit is set, white space data characters in the pattern are totally
! 646: ignored except when escaped or inside a character class. White space does not
1.1 misho 647: include the VT character (code 11). In addition, characters between an
648: unescaped # outside a character class and the next newline, inclusive, are also
649: ignored. This is equivalent to Perl's /x option, and it can be changed within a
650: pattern by a (?x) option setting.
651: .P
652: Which characters are interpreted as newlines is controlled by the options
653: passed to \fBpcre_compile()\fP or by a special sequence at the start of the
654: pattern, as described in the section entitled
655: .\" HTML <a href="pcrepattern.html#newlines">
656: .\" </a>
657: "Newline conventions"
658: .\"
659: in the \fBpcrepattern\fP documentation. Note that the end of this type of
660: comment is a literal newline sequence in the pattern; escape sequences that
661: happen to represent a newline do not count.
662: .P
663: This option makes it possible to include comments inside complicated patterns.
1.1.1.3 ! misho 664: Note, however, that this applies only to data characters. White space characters
1.1 misho 665: may never appear within special character sequences in a pattern, for example
666: within the sequence (?( that introduces a conditional subpattern.
667: .sp
668: PCRE_EXTRA
669: .sp
670: This option was invented in order to turn on additional functionality of PCRE
671: that is incompatible with Perl, but it is currently of very little use. When
672: set, any backslash in a pattern that is followed by a letter that has no
673: special meaning causes an error, thus reserving these combinations for future
674: expansion. By default, as in Perl, a backslash followed by a letter with no
675: special meaning is treated as a literal. (Perl can, however, be persuaded to
676: give an error for this, by running it with the -w option.) There are at present
677: no other features controlled by this option. It can also be set by a (?X)
678: option setting within a pattern.
679: .sp
680: PCRE_FIRSTLINE
681: .sp
682: If this option is set, an unanchored pattern is required to match before or at
683: the first newline in the subject string, though the matched text may continue
684: over the newline.
685: .sp
686: PCRE_JAVASCRIPT_COMPAT
687: .sp
688: If this option is set, PCRE's behaviour is changed in some ways so that it is
689: compatible with JavaScript rather than Perl. The changes are as follows:
690: .P
691: (1) A lone closing square bracket in a pattern causes a compile-time error,
692: because this is illegal in JavaScript (by default it is treated as a data
693: character). Thus, the pattern AB]CD becomes illegal when this option is set.
694: .P
695: (2) At run time, a back reference to an unset subpattern group matches an empty
696: string (by default this causes the current matching alternative to fail). A
697: pattern such as (\e1)(a) succeeds when this option is set (assuming it can find
698: an "a" in the subject), whereas it fails by default, for Perl compatibility.
699: .P
700: (3) \eU matches an upper case "U" character; by default \eU causes a compile
701: time error (Perl uses \eU to upper case subsequent characters).
702: .P
703: (4) \eu matches a lower case "u" character unless it is followed by four
704: hexadecimal digits, in which case the hexadecimal number defines the code point
705: to match. By default, \eu causes a compile time error (Perl uses it to upper
706: case the following character).
707: .P
708: (5) \ex matches a lower case "x" character unless it is followed by two
709: hexadecimal digits, in which case the hexadecimal number defines the code point
710: to match. By default, as in Perl, a hexadecimal number is always expected after
711: \ex, but it may have zero, one, or two digits (so, for example, \exz matches a
712: binary zero character followed by z).
713: .sp
714: PCRE_MULTILINE
715: .sp
716: By default, PCRE treats the subject string as consisting of a single line of
717: characters (even if it actually contains newlines). The "start of line"
718: metacharacter (^) matches only at the start of the string, while the "end of
719: line" metacharacter ($) matches only at the end of the string, or before a
720: terminating newline (unless PCRE_DOLLAR_ENDONLY is set). This is the same as
721: Perl.
722: .P
723: When PCRE_MULTILINE it is set, the "start of line" and "end of line" constructs
724: match immediately following or immediately before internal newlines in the
725: subject string, respectively, as well as at the very start and end. This is
726: equivalent to Perl's /m option, and it can be changed within a pattern by a
727: (?m) option setting. If there are no newlines in a subject string, or no
728: occurrences of ^ or $ in a pattern, setting PCRE_MULTILINE has no effect.
729: .sp
730: PCRE_NEWLINE_CR
731: PCRE_NEWLINE_LF
732: PCRE_NEWLINE_CRLF
733: PCRE_NEWLINE_ANYCRLF
734: PCRE_NEWLINE_ANY
735: .sp
736: These options override the default newline definition that was chosen when PCRE
737: was built. Setting the first or the second specifies that a newline is
738: indicated by a single character (CR or LF, respectively). Setting
739: PCRE_NEWLINE_CRLF specifies that a newline is indicated by the two-character
740: CRLF sequence. Setting PCRE_NEWLINE_ANYCRLF specifies that any of the three
741: preceding sequences should be recognized. Setting PCRE_NEWLINE_ANY specifies
742: that any Unicode newline sequence should be recognized. The Unicode newline
743: sequences are the three just mentioned, plus the single characters VT (vertical
1.1.1.3 ! misho 744: tab, U+000B), FF (form feed, U+000C), NEL (next line, U+0085), LS (line
1.1.1.2 misho 745: separator, U+2028), and PS (paragraph separator, U+2029). For the 8-bit
746: library, the last two are recognized only in UTF-8 mode.
1.1 misho 747: .P
748: The newline setting in the options word uses three bits that are treated
749: as a number, giving eight possibilities. Currently only six are used (default
750: plus the five values above). This means that if you set more than one newline
751: option, the combination may or may not be sensible. For example,
752: PCRE_NEWLINE_CR with PCRE_NEWLINE_LF is equivalent to PCRE_NEWLINE_CRLF, but
753: other combinations may yield unused numbers and cause an error.
754: .P
755: The only time that a line break in a pattern is specially recognized when
1.1.1.3 ! misho 756: compiling is when PCRE_EXTENDED is set. CR and LF are white space characters,
1.1 misho 757: and so are ignored in this mode. Also, an unescaped # outside a character class
758: indicates a comment that lasts until after the next line break sequence. In
759: other circumstances, line break sequences in patterns are treated as literal
760: data.
761: .P
762: The newline option that is set at compile time becomes the default that is used
763: for \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP, but it can be overridden.
764: .sp
765: PCRE_NO_AUTO_CAPTURE
766: .sp
767: If this option is set, it disables the use of numbered capturing parentheses in
768: the pattern. Any opening parenthesis that is not followed by ? behaves as if it
769: were followed by ?: but named parentheses can still be used for capturing (and
770: they acquire numbers in the usual way). There is no equivalent of this option
771: in Perl.
772: .sp
773: NO_START_OPTIMIZE
774: .sp
775: This is an option that acts at matching time; that is, it is really an option
776: for \fBpcre_exec()\fP or \fBpcre_dfa_exec()\fP. If it is set at compile time,
777: it is remembered with the compiled pattern and assumed at matching time. For
778: details see the discussion of PCRE_NO_START_OPTIMIZE
779: .\" HTML <a href="#execoptions">
780: .\" </a>
781: below.
782: .\"
783: .sp
784: PCRE_UCP
785: .sp
786: This option changes the way PCRE processes \eB, \eb, \eD, \ed, \eS, \es, \eW,
787: \ew, and some of the POSIX character classes. By default, only ASCII characters
788: are recognized, but if PCRE_UCP is set, Unicode properties are used instead to
789: classify characters. More details are given in the section on
790: .\" HTML <a href="pcre.html#genericchartypes">
791: .\" </a>
792: generic character types
793: .\"
794: in the
795: .\" HREF
796: \fBpcrepattern\fP
797: .\"
798: page. If you set PCRE_UCP, matching one of the items it affects takes much
799: longer. The option is available only if PCRE has been compiled with Unicode
800: property support.
801: .sp
802: PCRE_UNGREEDY
803: .sp
804: This option inverts the "greediness" of the quantifiers so that they are not
805: greedy by default, but become greedy if followed by "?". It is not compatible
806: with Perl. It can also be set by a (?U) option setting within the pattern.
807: .sp
808: PCRE_UTF8
809: .sp
810: This option causes PCRE to regard both the pattern and the subject as strings
1.1.1.2 misho 811: of UTF-8 characters instead of single-byte strings. However, it is available
812: only when PCRE is built to include UTF support. If not, the use of this option
813: provokes an error. Details of how this option changes the behaviour of PCRE are
814: given in the
1.1 misho 815: .\" HREF
816: \fBpcreunicode\fP
817: .\"
818: page.
819: .sp
820: PCRE_NO_UTF8_CHECK
821: .sp
1.1.1.2 misho 822: When PCRE_UTF8 is set, the validity of the pattern as a UTF-8
823: string is automatically checked. There is a discussion about the
824: .\" HTML <a href="pcreunicode.html#utf8strings">
1.1 misho 825: .\" </a>
826: validity of UTF-8 strings
827: .\"
1.1.1.2 misho 828: in the
1.1 misho 829: .\" HREF
1.1.1.2 misho 830: \fBpcreunicode\fP
1.1 misho 831: .\"
1.1.1.2 misho 832: page. If an invalid UTF-8 sequence is found, \fBpcre_compile()\fP returns an
833: error. If you already know that your pattern is valid, and you want to skip
834: this check for performance reasons, you can set the PCRE_NO_UTF8_CHECK option.
835: When it is set, the effect of passing an invalid UTF-8 string as a pattern is
836: undefined. It may cause your program to crash. Note that this option can also
837: be passed to \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP, to suppress the
838: validity checking of subject strings.
1.1 misho 839: .
840: .
841: .SH "COMPILATION ERROR CODES"
842: .rs
843: .sp
844: The following table lists the error codes than may be returned by
845: \fBpcre_compile2()\fP, along with the error messages that may be returned by
1.1.1.2 misho 846: both compiling functions. Note that error messages are always 8-bit ASCII
847: strings, even in 16-bit mode. As PCRE has developed, some error codes have
848: fallen out of use. To avoid confusion, they have not been re-used.
1.1 misho 849: .sp
850: 0 no error
851: 1 \e at end of pattern
852: 2 \ec at end of pattern
853: 3 unrecognized character follows \e
854: 4 numbers out of order in {} quantifier
855: 5 number too big in {} quantifier
856: 6 missing terminating ] for character class
857: 7 invalid escape sequence in character class
858: 8 range out of order in character class
859: 9 nothing to repeat
860: 10 [this code is not in use]
861: 11 internal error: unexpected repeat
862: 12 unrecognized character after (? or (?-
863: 13 POSIX named classes are supported only within a class
864: 14 missing )
865: 15 reference to non-existent subpattern
866: 16 erroffset passed as NULL
867: 17 unknown option bit(s) set
868: 18 missing ) after comment
869: 19 [this code is not in use]
870: 20 regular expression is too large
871: 21 failed to get memory
872: 22 unmatched parentheses
873: 23 internal error: code overflow
874: 24 unrecognized character after (?<
875: 25 lookbehind assertion is not fixed length
876: 26 malformed number or name after (?(
877: 27 conditional group contains more than two branches
878: 28 assertion expected after (?(
879: 29 (?R or (?[+-]digits must be followed by )
880: 30 unknown POSIX class name
881: 31 POSIX collating elements are not supported
1.1.1.2 misho 882: 32 this version of PCRE is compiled without UTF support
1.1 misho 883: 33 [this code is not in use]
884: 34 character value in \ex{...} sequence is too large
885: 35 invalid condition (?(0)
886: 36 \eC not allowed in lookbehind assertion
887: 37 PCRE does not support \eL, \el, \eN{name}, \eU, or \eu
888: 38 number after (?C is > 255
889: 39 closing ) for (?C expected
890: 40 recursive call could loop indefinitely
891: 41 unrecognized character after (?P
892: 42 syntax error in subpattern name (missing terminator)
893: 43 two named subpatterns have the same name
1.1.1.2 misho 894: 44 invalid UTF-8 string (specifically UTF-8)
1.1 misho 895: 45 support for \eP, \ep, and \eX has not been compiled
896: 46 malformed \eP or \ep sequence
897: 47 unknown property name after \eP or \ep
898: 48 subpattern name is too long (maximum 32 characters)
899: 49 too many named subpatterns (maximum 10000)
900: 50 [this code is not in use]
1.1.1.2 misho 901: 51 octal value is greater than \e377 in 8-bit non-UTF-8 mode
1.1 misho 902: 52 internal error: overran compiling workspace
903: 53 internal error: previously-checked referenced subpattern
904: not found
905: 54 DEFINE group contains more than one branch
906: 55 repeating a DEFINE group is not allowed
907: 56 inconsistent NEWLINE options
908: 57 \eg is not followed by a braced, angle-bracketed, or quoted
909: name/number or by a plain number
910: 58 a numbered reference must not be zero
911: 59 an argument is not allowed for (*ACCEPT), (*FAIL), or (*COMMIT)
912: 60 (*VERB) not recognized
913: 61 number is too big
914: 62 subpattern name expected
915: 63 digit expected after (?+
916: 64 ] is an invalid data character in JavaScript compatibility mode
917: 65 different names for subpatterns of the same number are
918: not allowed
919: 66 (*MARK) must have an argument
1.1.1.2 misho 920: 67 this version of PCRE is not compiled with Unicode property
921: support
1.1 misho 922: 68 \ec must be followed by an ASCII character
923: 69 \ek is not followed by a braced, angle-bracketed, or quoted name
1.1.1.2 misho 924: 70 internal error: unknown opcode in find_fixedlength()
925: 71 \eN is not supported in a class
926: 72 too many forward references
927: 73 disallowed Unicode code point (>= 0xd800 && <= 0xdfff)
928: 74 invalid UTF-16 string (specifically UTF-16)
1.1.1.3 ! misho 929: 75 name is too long in (*MARK), (*PRUNE), (*SKIP), or (*THEN)
! 930: 76 character value in \eu.... sequence is too large
1.1 misho 931: .sp
932: The numbers 32 and 10000 in errors 48 and 49 are defaults; different values may
933: be used if the limits were changed when PCRE was built.
934: .
935: .
936: .\" HTML <a name="studyingapattern"></a>
937: .SH "STUDYING A PATTERN"
938: .rs
939: .sp
940: .B pcre_extra *pcre_study(const pcre *\fIcode\fP, int \fIoptions\fP
941: .ti +5n
942: .B const char **\fIerrptr\fP);
943: .PP
944: If a compiled pattern is going to be used several times, it is worth spending
945: more time analyzing it in order to speed up the time taken for matching. The
946: function \fBpcre_study()\fP takes a pointer to a compiled pattern as its first
947: argument. If studying the pattern produces additional information that will
948: help speed up matching, \fBpcre_study()\fP returns a pointer to a
949: \fBpcre_extra\fP block, in which the \fIstudy_data\fP field points to the
950: results of the study.
951: .P
952: The returned value from \fBpcre_study()\fP can be passed directly to
953: \fBpcre_exec()\fP or \fBpcre_dfa_exec()\fP. However, a \fBpcre_extra\fP block
954: also contains other fields that can be set by the caller before the block is
955: passed; these are described
956: .\" HTML <a href="#extradata">
957: .\" </a>
958: below
959: .\"
960: in the section on matching a pattern.
961: .P
962: If studying the pattern does not produce any useful information,
963: \fBpcre_study()\fP returns NULL. In that circumstance, if the calling program
964: wants to pass any of the other fields to \fBpcre_exec()\fP or
965: \fBpcre_dfa_exec()\fP, it must set up its own \fBpcre_extra\fP block.
966: .P
1.1.1.3 ! misho 967: The second argument of \fBpcre_study()\fP contains option bits. There are three
! 968: options:
! 969: .sp
! 970: PCRE_STUDY_JIT_COMPILE
! 971: PCRE_STUDY_JIT_PARTIAL_HARD_COMPILE
! 972: PCRE_STUDY_JIT_PARTIAL_SOFT_COMPILE
! 973: .sp
! 974: If any of these are set, and the just-in-time compiler is available, the
! 975: pattern is further compiled into machine code that executes much faster than
! 976: the \fBpcre_exec()\fP interpretive matching function. If the just-in-time
! 977: compiler is not available, these options are ignored. All other bits in the
! 978: \fIoptions\fP argument must be zero.
1.1 misho 979: .P
980: JIT compilation is a heavyweight optimization. It can take some time for
981: patterns to be analyzed, and for one-off matches and simple patterns the
982: benefit of faster execution might be offset by a much slower study time.
983: Not all patterns can be optimized by the JIT compiler. For those that cannot be
984: handled, matching automatically falls back to the \fBpcre_exec()\fP
985: interpreter. For more details, see the
986: .\" HREF
987: \fBpcrejit\fP
988: .\"
989: documentation.
990: .P
991: The third argument for \fBpcre_study()\fP is a pointer for an error message. If
992: studying succeeds (even if no data is returned), the variable it points to is
993: set to NULL. Otherwise it is set to point to a textual error message. This is a
994: static string that is part of the library. You must not try to free it. You
995: should test the error pointer for NULL after calling \fBpcre_study()\fP, to be
996: sure that it has run successfully.
997: .P
998: When you are finished with a pattern, you can free the memory used for the
999: study data by calling \fBpcre_free_study()\fP. This function was added to the
1000: API for release 8.20. For earlier versions, the memory could be freed with
1001: \fBpcre_free()\fP, just like the pattern itself. This will still work in cases
1.1.1.3 ! misho 1002: where JIT optimization is not used, but it is advisable to change to the new
! 1003: function when convenient.
1.1 misho 1004: .P
1005: This is a typical way in which \fBpcre_study\fP() is used (except that in a
1006: real application there should be tests for errors):
1007: .sp
1008: int rc;
1009: pcre *re;
1010: pcre_extra *sd;
1011: re = pcre_compile("pattern", 0, &error, &erroroffset, NULL);
1012: sd = pcre_study(
1013: re, /* result of pcre_compile() */
1014: 0, /* no options */
1015: &error); /* set to NULL or points to a message */
1016: rc = pcre_exec( /* see below for details of pcre_exec() options */
1017: re, sd, "subject", 7, 0, 0, ovector, 30);
1018: ...
1019: pcre_free_study(sd);
1020: pcre_free(re);
1021: .sp
1022: Studying a pattern does two things: first, a lower bound for the length of
1023: subject string that is needed to match the pattern is computed. This does not
1024: mean that there are any strings of that length that match, but it does
1025: guarantee that no shorter strings match. The value is used by
1026: \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP to avoid wasting time by trying to
1027: match strings that are shorter than the lower bound. You can find out the value
1028: in a calling program via the \fBpcre_fullinfo()\fP function.
1029: .P
1030: Studying a pattern is also useful for non-anchored patterns that do not have a
1031: single fixed starting character. A bitmap of possible starting bytes is
1032: created. This speeds up finding a position in the subject at which to start
1.1.1.2 misho 1033: matching. (In 16-bit mode, the bitmap is used for 16-bit values less than 256.)
1.1 misho 1034: .P
1035: These two optimizations apply to both \fBpcre_exec()\fP and
1.1.1.3 ! misho 1036: \fBpcre_dfa_exec()\fP, and the information is also used by the JIT compiler.
! 1037: The optimizations can be disabled by setting the PCRE_NO_START_OPTIMIZE option
! 1038: when calling \fBpcre_exec()\fP or \fBpcre_dfa_exec()\fP, but if this is done,
! 1039: JIT execution is also disabled. You might want to do this if your pattern
! 1040: contains callouts or (*MARK) and you want to make use of these facilities in
! 1041: cases where matching fails. See the discussion of PCRE_NO_START_OPTIMIZE
1.1 misho 1042: .\" HTML <a href="#execoptions">
1043: .\" </a>
1044: below.
1045: .\"
1046: .
1047: .
1048: .\" HTML <a name="localesupport"></a>
1049: .SH "LOCALE SUPPORT"
1050: .rs
1051: .sp
1052: PCRE handles caseless matching, and determines whether characters are letters,
1053: digits, or whatever, by reference to a set of tables, indexed by character
1.1.1.2 misho 1054: value. When running in UTF-8 mode, this applies only to characters
1055: with codes less than 128. By default, higher-valued codes never match escapes
1056: such as \ew or \ed, but they can be tested with \ep if PCRE is built with
1057: Unicode character property support. Alternatively, the PCRE_UCP option can be
1058: set at compile time; this causes \ew and friends to use Unicode property
1059: support instead of built-in tables. The use of locales with Unicode is
1060: discouraged. If you are handling characters with codes greater than 128, you
1061: should either use UTF-8 and Unicode, or use locales, but not try to mix the
1062: two.
1.1 misho 1063: .P
1064: PCRE contains an internal set of tables that are used when the final argument
1065: of \fBpcre_compile()\fP is NULL. These are sufficient for many applications.
1066: Normally, the internal tables recognize only ASCII characters. However, when
1067: PCRE is built, it is possible to cause the internal tables to be rebuilt in the
1068: default "C" locale of the local system, which may cause them to be different.
1069: .P
1070: The internal tables can always be overridden by tables supplied by the
1071: application that calls PCRE. These may be created in a different locale from
1072: the default. As more and more applications change to using Unicode, the need
1073: for this locale support is expected to die away.
1074: .P
1075: External tables are built by calling the \fBpcre_maketables()\fP function,
1076: which has no arguments, in the relevant locale. The result can then be passed
1077: to \fBpcre_compile()\fP or \fBpcre_exec()\fP as often as necessary. For
1078: example, to build and use tables that are appropriate for the French locale
1079: (where accented characters with values greater than 128 are treated as letters),
1080: the following code could be used:
1081: .sp
1082: setlocale(LC_CTYPE, "fr_FR");
1083: tables = pcre_maketables();
1084: re = pcre_compile(..., tables);
1085: .sp
1086: The locale name "fr_FR" is used on Linux and other Unix-like systems; if you
1087: are using Windows, the name for the French locale is "french".
1088: .P
1089: When \fBpcre_maketables()\fP runs, the tables are built in memory that is
1090: obtained via \fBpcre_malloc\fP. It is the caller's responsibility to ensure
1091: that the memory containing the tables remains available for as long as it is
1092: needed.
1093: .P
1094: The pointer that is passed to \fBpcre_compile()\fP is saved with the compiled
1095: pattern, and the same tables are used via this pointer by \fBpcre_study()\fP
1096: and normally also by \fBpcre_exec()\fP. Thus, by default, for any single
1097: pattern, compilation, studying and matching all happen in the same locale, but
1098: different patterns can be compiled in different locales.
1099: .P
1100: It is possible to pass a table pointer or NULL (indicating the use of the
1101: internal tables) to \fBpcre_exec()\fP. Although not intended for this purpose,
1102: this facility could be used to match a pattern in a different locale from the
1103: one in which it was compiled. Passing table pointers at run time is discussed
1104: below in the section on matching a pattern.
1105: .
1106: .
1107: .\" HTML <a name="infoaboutpattern"></a>
1108: .SH "INFORMATION ABOUT A PATTERN"
1109: .rs
1110: .sp
1111: .B int pcre_fullinfo(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
1112: .ti +5n
1113: .B int \fIwhat\fP, void *\fIwhere\fP);
1114: .PP
1115: The \fBpcre_fullinfo()\fP function returns information about a compiled
1.1.1.2 misho 1116: pattern. It replaces the \fBpcre_info()\fP function, which was removed from the
1117: library at version 8.30, after more than 10 years of obsolescence.
1.1 misho 1118: .P
1119: The first argument for \fBpcre_fullinfo()\fP is a pointer to the compiled
1120: pattern. The second argument is the result of \fBpcre_study()\fP, or NULL if
1121: the pattern was not studied. The third argument specifies which piece of
1122: information is required, and the fourth argument is a pointer to a variable
1123: to receive the data. The yield of the function is zero for success, or one of
1124: the following negative numbers:
1125: .sp
1.1.1.2 misho 1126: PCRE_ERROR_NULL the argument \fIcode\fP was NULL
1127: the argument \fIwhere\fP was NULL
1128: PCRE_ERROR_BADMAGIC the "magic number" was not found
1129: PCRE_ERROR_BADENDIANNESS the pattern was compiled with different
1130: endianness
1131: PCRE_ERROR_BADOPTION the value of \fIwhat\fP was invalid
1.1 misho 1132: .sp
1133: The "magic number" is placed at the start of each compiled pattern as an simple
1.1.1.2 misho 1134: check against passing an arbitrary memory pointer. The endianness error can
1135: occur if a compiled pattern is saved and reloaded on a different host. Here is
1136: a typical call of \fBpcre_fullinfo()\fP, to obtain the length of the compiled
1137: pattern:
1.1 misho 1138: .sp
1139: int rc;
1140: size_t length;
1141: rc = pcre_fullinfo(
1142: re, /* result of pcre_compile() */
1143: sd, /* result of pcre_study(), or NULL */
1144: PCRE_INFO_SIZE, /* what is required */
1145: &length); /* where to put the data */
1146: .sp
1147: The possible values for the third argument are defined in \fBpcre.h\fP, and are
1148: as follows:
1149: .sp
1150: PCRE_INFO_BACKREFMAX
1151: .sp
1152: Return the number of the highest back reference in the pattern. The fourth
1153: argument should point to an \fBint\fP variable. Zero is returned if there are
1154: no back references.
1155: .sp
1156: PCRE_INFO_CAPTURECOUNT
1157: .sp
1158: Return the number of capturing subpatterns in the pattern. The fourth argument
1159: should point to an \fBint\fP variable.
1160: .sp
1161: PCRE_INFO_DEFAULT_TABLES
1162: .sp
1163: Return a pointer to the internal default character tables within PCRE. The
1164: fourth argument should point to an \fBunsigned char *\fP variable. This
1165: information call is provided for internal use by the \fBpcre_study()\fP
1166: function. External callers can cause PCRE to use its internal tables by passing
1167: a NULL table pointer.
1168: .sp
1169: PCRE_INFO_FIRSTBYTE
1170: .sp
1.1.1.2 misho 1171: Return information about the first data unit of any matched string, for a
1172: non-anchored pattern. (The name of this option refers to the 8-bit library,
1173: where data units are bytes.) The fourth argument should point to an \fBint\fP
1174: variable.
1175: .P
1176: If there is a fixed first value, for example, the letter "c" from a pattern
1177: such as (cat|cow|coyote), its value is returned. In the 8-bit library, the
1178: value is always less than 256; in the 16-bit library the value can be up to
1179: 0xffff.
1.1 misho 1180: .P
1.1.1.2 misho 1181: If there is no fixed first value, and if either
1.1 misho 1182: .sp
1183: (a) the pattern was compiled with the PCRE_MULTILINE option, and every branch
1184: starts with "^", or
1185: .sp
1186: (b) every branch of the pattern starts with ".*" and PCRE_DOTALL is not set
1187: (if it were set, the pattern would be anchored),
1188: .sp
1189: -1 is returned, indicating that the pattern matches only at the start of a
1190: subject string or after any newline within the string. Otherwise -2 is
1191: returned. For anchored patterns, -2 is returned.
1192: .sp
1193: PCRE_INFO_FIRSTTABLE
1194: .sp
1195: If the pattern was studied, and this resulted in the construction of a 256-bit
1.1.1.2 misho 1196: table indicating a fixed set of values for the first data unit in any matching
1.1 misho 1197: string, a pointer to the table is returned. Otherwise NULL is returned. The
1198: fourth argument should point to an \fBunsigned char *\fP variable.
1199: .sp
1200: PCRE_INFO_HASCRORLF
1201: .sp
1202: Return 1 if the pattern contains any explicit matches for CR or LF characters,
1203: otherwise 0. The fourth argument should point to an \fBint\fP variable. An
1204: explicit match is either a literal CR or LF character, or \er or \en.
1205: .sp
1206: PCRE_INFO_JCHANGED
1207: .sp
1208: Return 1 if the (?J) or (?-J) option setting is used in the pattern, otherwise
1209: 0. The fourth argument should point to an \fBint\fP variable. (?J) and
1210: (?-J) set and unset the local PCRE_DUPNAMES option, respectively.
1211: .sp
1212: PCRE_INFO_JIT
1213: .sp
1.1.1.3 ! misho 1214: Return 1 if the pattern was studied with one of the JIT options, and
1.1 misho 1215: just-in-time compiling was successful. The fourth argument should point to an
1216: \fBint\fP variable. A return value of 0 means that JIT support is not available
1.1.1.3 ! misho 1217: in this version of PCRE, or that the pattern was not studied with a JIT option,
! 1218: or that the JIT compiler could not handle this particular pattern. See the
1.1 misho 1219: .\" HREF
1220: \fBpcrejit\fP
1221: .\"
1222: documentation for details of what can and cannot be handled.
1223: .sp
1224: PCRE_INFO_JITSIZE
1225: .sp
1.1.1.3 ! misho 1226: If the pattern was successfully studied with a JIT option, return the size of
! 1227: the JIT compiled code, otherwise return zero. The fourth argument should point
! 1228: to a \fBsize_t\fP variable.
1.1 misho 1229: .sp
1230: PCRE_INFO_LASTLITERAL
1231: .sp
1.1.1.2 misho 1232: Return the value of the rightmost literal data unit that must exist in any
1233: matched string, other than at its start, if such a value has been recorded. The
1234: fourth argument should point to an \fBint\fP variable. If there is no such
1235: value, -1 is returned. For anchored patterns, a last literal value is recorded
1236: only if it follows something of variable length. For example, for the pattern
1.1 misho 1237: /^a\ed+z\ed+/ the returned value is "z", but for /^a\edz\ed/ the returned value
1238: is -1.
1239: .sp
1.1.1.3 ! misho 1240: PCRE_INFO_MAXLOOKBEHIND
! 1241: .sp
! 1242: Return the number of characters (NB not bytes) in the longest lookbehind
! 1243: assertion in the pattern. Note that the simple assertions \eb and \eB require a
! 1244: one-character lookbehind. This information is useful when doing multi-segment
! 1245: matching using the partial matching facilities.
! 1246: .sp
1.1 misho 1247: PCRE_INFO_MINLENGTH
1248: .sp
1249: If the pattern was studied and a minimum length for matching subject strings
1250: was computed, its value is returned. Otherwise the returned value is -1. The
1.1.1.2 misho 1251: value is a number of characters, which in UTF-8 mode may be different from the
1252: number of bytes. The fourth argument should point to an \fBint\fP variable. A
1.1 misho 1253: non-negative value is a lower bound to the length of any matching string. There
1254: may not be any strings of that length that do actually match, but every string
1255: that does match is at least that long.
1256: .sp
1257: PCRE_INFO_NAMECOUNT
1258: PCRE_INFO_NAMEENTRYSIZE
1259: PCRE_INFO_NAMETABLE
1260: .sp
1261: PCRE supports the use of named as well as numbered capturing parentheses. The
1262: names are just an additional way of identifying the parentheses, which still
1263: acquire numbers. Several convenience functions such as
1264: \fBpcre_get_named_substring()\fP are provided for extracting captured
1265: substrings by name. It is also possible to extract the data directly, by first
1266: converting the name to a number in order to access the correct pointers in the
1267: output vector (described with \fBpcre_exec()\fP below). To do the conversion,
1268: you need to use the name-to-number map, which is described by these three
1269: values.
1270: .P
1271: The map consists of a number of fixed-size entries. PCRE_INFO_NAMECOUNT gives
1272: the number of entries, and PCRE_INFO_NAMEENTRYSIZE gives the size of each
1273: entry; both of these return an \fBint\fP value. The entry size depends on the
1274: length of the longest name. PCRE_INFO_NAMETABLE returns a pointer to the first
1.1.1.2 misho 1275: entry of the table. This is a pointer to \fBchar\fP in the 8-bit library, where
1276: the first two bytes of each entry are the number of the capturing parenthesis,
1277: most significant byte first. In the 16-bit library, the pointer points to
1278: 16-bit data units, the first of which contains the parenthesis number. The rest
1279: of the entry is the corresponding name, zero terminated.
1.1 misho 1280: .P
1281: The names are in alphabetical order. Duplicate names may appear if (?| is used
1282: to create multiple groups with the same number, as described in the
1283: .\" HTML <a href="pcrepattern.html#dupsubpatternnumber">
1284: .\" </a>
1285: section on duplicate subpattern numbers
1286: .\"
1287: in the
1288: .\" HREF
1289: \fBpcrepattern\fP
1290: .\"
1291: page. Duplicate names for subpatterns with different numbers are permitted only
1292: if PCRE_DUPNAMES is set. In all cases of duplicate names, they appear in the
1293: table in the order in which they were found in the pattern. In the absence of
1294: (?| this is the order of increasing number; when (?| is used this is not
1295: necessarily the case because later subpatterns may have lower numbers.
1296: .P
1297: As a simple example of the name/number table, consider the following pattern
1.1.1.2 misho 1298: after compilation by the 8-bit library (assume PCRE_EXTENDED is set, so white
1299: space - including newlines - is ignored):
1.1 misho 1300: .sp
1301: .\" JOIN
1302: (?<date> (?<year>(\ed\ed)?\ed\ed) -
1303: (?<month>\ed\ed) - (?<day>\ed\ed) )
1304: .sp
1305: There are four named subpatterns, so the table has four entries, and each entry
1306: in the table is eight bytes long. The table is as follows, with non-printing
1307: bytes shows in hexadecimal, and undefined bytes shown as ??:
1308: .sp
1309: 00 01 d a t e 00 ??
1310: 00 05 d a y 00 ?? ??
1311: 00 04 m o n t h 00
1312: 00 02 y e a r 00 ??
1313: .sp
1314: When writing code to extract data from named subpatterns using the
1315: name-to-number map, remember that the length of the entries is likely to be
1316: different for each compiled pattern.
1317: .sp
1318: PCRE_INFO_OKPARTIAL
1319: .sp
1320: Return 1 if the pattern can be used for partial matching with
1321: \fBpcre_exec()\fP, otherwise 0. The fourth argument should point to an
1322: \fBint\fP variable. From release 8.00, this always returns 1, because the
1323: restrictions that previously applied to partial matching have been lifted. The
1324: .\" HREF
1325: \fBpcrepartial\fP
1326: .\"
1327: documentation gives details of partial matching.
1328: .sp
1329: PCRE_INFO_OPTIONS
1330: .sp
1331: Return a copy of the options with which the pattern was compiled. The fourth
1332: argument should point to an \fBunsigned long int\fP variable. These option bits
1333: are those specified in the call to \fBpcre_compile()\fP, modified by any
1334: top-level option settings at the start of the pattern itself. In other words,
1335: they are the options that will be in force when matching starts. For example,
1336: if the pattern /(?im)abc(?-i)d/ is compiled with the PCRE_EXTENDED option, the
1337: result is PCRE_CASELESS, PCRE_MULTILINE, and PCRE_EXTENDED.
1338: .P
1339: A pattern is automatically anchored by PCRE if all of its top-level
1340: alternatives begin with one of the following:
1341: .sp
1342: ^ unless PCRE_MULTILINE is set
1343: \eA always
1344: \eG always
1345: .\" JOIN
1346: .* if PCRE_DOTALL is set and there are no back
1347: references to the subpattern in which .* appears
1348: .sp
1349: For such patterns, the PCRE_ANCHORED bit is set in the options returned by
1350: \fBpcre_fullinfo()\fP.
1351: .sp
1352: PCRE_INFO_SIZE
1353: .sp
1.1.1.2 misho 1354: Return the size of the compiled pattern in bytes (for both libraries). The
1355: fourth argument should point to a \fBsize_t\fP variable. This value does not
1356: include the size of the \fBpcre\fP structure that is returned by
1357: \fBpcre_compile()\fP. The value that is passed as the argument to
1358: \fBpcre_malloc()\fP when \fBpcre_compile()\fP is getting memory in which to
1359: place the compiled data is the value returned by this option plus the size of
1360: the \fBpcre\fP structure. Studying a compiled pattern, with or without JIT,
1361: does not alter the value returned by this option.
1.1 misho 1362: .sp
1363: PCRE_INFO_STUDYSIZE
1364: .sp
1.1.1.2 misho 1365: Return the size in bytes of the data block pointed to by the \fIstudy_data\fP
1366: field in a \fBpcre_extra\fP block. If \fBpcre_extra\fP is NULL, or there is no
1367: study data, zero is returned. The fourth argument should point to a
1368: \fBsize_t\fP variable. The \fIstudy_data\fP field is set by \fBpcre_study()\fP
1369: to record information that will speed up matching (see the section entitled
1.1 misho 1370: .\" HTML <a href="#studyingapattern">
1371: .\" </a>
1372: "Studying a pattern"
1373: .\"
1374: above). The format of the \fIstudy_data\fP block is private, but its length
1375: is made available via this option so that it can be saved and restored (see the
1376: .\" HREF
1377: \fBpcreprecompile\fP
1378: .\"
1379: documentation for details).
1380: .
1381: .
1382: .SH "REFERENCE COUNTS"
1383: .rs
1384: .sp
1385: .B int pcre_refcount(pcre *\fIcode\fP, int \fIadjust\fP);
1386: .PP
1387: The \fBpcre_refcount()\fP function is used to maintain a reference count in the
1388: data block that contains a compiled pattern. It is provided for the benefit of
1389: applications that operate in an object-oriented manner, where different parts
1390: of the application may be using the same compiled pattern, but you want to free
1391: the block when they are all done.
1392: .P
1393: When a pattern is compiled, the reference count field is initialized to zero.
1394: It is changed only by calling this function, whose action is to add the
1395: \fIadjust\fP value (which may be positive or negative) to it. The yield of the
1396: function is the new value. However, the value of the count is constrained to
1397: lie between 0 and 65535, inclusive. If the new value is outside these limits,
1398: it is forced to the appropriate limit value.
1399: .P
1400: Except when it is zero, the reference count is not correctly preserved if a
1401: pattern is compiled on one host and then transferred to a host whose byte-order
1402: is different. (This seems a highly unlikely scenario.)
1403: .
1404: .
1405: .SH "MATCHING A PATTERN: THE TRADITIONAL FUNCTION"
1406: .rs
1407: .sp
1408: .B int pcre_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
1409: .ti +5n
1410: .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
1411: .ti +5n
1412: .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP);
1413: .P
1414: The function \fBpcre_exec()\fP is called to match a subject string against a
1415: compiled pattern, which is passed in the \fIcode\fP argument. If the
1416: pattern was studied, the result of the study should be passed in the
1417: \fIextra\fP argument. You can call \fBpcre_exec()\fP with the same \fIcode\fP
1418: and \fIextra\fP arguments as many times as you like, in order to match
1419: different subject strings with the same pattern.
1420: .P
1421: This function is the main matching facility of the library, and it operates in
1422: a Perl-like manner. For specialist use there is also an alternative matching
1423: function, which is described
1424: .\" HTML <a href="#dfamatch">
1425: .\" </a>
1426: below
1427: .\"
1428: in the section about the \fBpcre_dfa_exec()\fP function.
1429: .P
1430: In most applications, the pattern will have been compiled (and optionally
1431: studied) in the same process that calls \fBpcre_exec()\fP. However, it is
1432: possible to save compiled patterns and study data, and then use them later
1433: in different processes, possibly even on different hosts. For a discussion
1434: about this, see the
1435: .\" HREF
1436: \fBpcreprecompile\fP
1437: .\"
1438: documentation.
1439: .P
1440: Here is an example of a simple call to \fBpcre_exec()\fP:
1441: .sp
1442: int rc;
1443: int ovector[30];
1444: rc = pcre_exec(
1445: re, /* result of pcre_compile() */
1446: NULL, /* we didn't study the pattern */
1447: "some string", /* the subject string */
1448: 11, /* the length of the subject string */
1449: 0, /* start at offset 0 in the subject */
1450: 0, /* default options */
1451: ovector, /* vector of integers for substring information */
1452: 30); /* number of elements (NOT size in bytes) */
1453: .
1454: .
1455: .\" HTML <a name="extradata"></a>
1456: .SS "Extra data for \fBpcre_exec()\fR"
1457: .rs
1458: .sp
1459: If the \fIextra\fP argument is not NULL, it must point to a \fBpcre_extra\fP
1460: data block. The \fBpcre_study()\fP function returns such a block (when it
1461: doesn't return NULL), but you can also create one for yourself, and pass
1462: additional information in it. The \fBpcre_extra\fP block contains the following
1463: fields (not necessarily in this order):
1464: .sp
1465: unsigned long int \fIflags\fP;
1466: void *\fIstudy_data\fP;
1467: void *\fIexecutable_jit\fP;
1468: unsigned long int \fImatch_limit\fP;
1469: unsigned long int \fImatch_limit_recursion\fP;
1470: void *\fIcallout_data\fP;
1471: const unsigned char *\fItables\fP;
1472: unsigned char **\fImark\fP;
1473: .sp
1.1.1.2 misho 1474: In the 16-bit version of this structure, the \fImark\fP field has type
1475: "PCRE_UCHAR16 **".
1476: .P
1.1.1.3 ! misho 1477: The \fIflags\fP field is used to specify which of the other fields are set. The
! 1478: flag bits are:
1.1 misho 1479: .sp
1.1.1.3 ! misho 1480: PCRE_EXTRA_CALLOUT_DATA
1.1 misho 1481: PCRE_EXTRA_EXECUTABLE_JIT
1.1.1.3 ! misho 1482: PCRE_EXTRA_MARK
1.1 misho 1483: PCRE_EXTRA_MATCH_LIMIT
1484: PCRE_EXTRA_MATCH_LIMIT_RECURSION
1.1.1.3 ! misho 1485: PCRE_EXTRA_STUDY_DATA
1.1 misho 1486: PCRE_EXTRA_TABLES
1487: .sp
1488: Other flag bits should be set to zero. The \fIstudy_data\fP field and sometimes
1489: the \fIexecutable_jit\fP field are set in the \fBpcre_extra\fP block that is
1490: returned by \fBpcre_study()\fP, together with the appropriate flag bits. You
1.1.1.3 ! misho 1491: should not set these yourself, but you may add to the block by setting other
! 1492: fields and their corresponding flag bits.
1.1 misho 1493: .P
1494: The \fImatch_limit\fP field provides a means of preventing PCRE from using up a
1495: vast amount of resources when running patterns that are not going to match,
1496: but which have a very large number of possibilities in their search trees. The
1497: classic example is a pattern that uses nested unlimited repeats.
1498: .P
1499: Internally, \fBpcre_exec()\fP uses a function called \fBmatch()\fP, which it
1500: calls repeatedly (sometimes recursively). The limit set by \fImatch_limit\fP is
1501: imposed on the number of times this function is called during a match, which
1502: has the effect of limiting the amount of backtracking that can take place. For
1503: patterns that are not anchored, the count restarts from zero for each position
1504: in the subject string.
1505: .P
1506: When \fBpcre_exec()\fP is called with a pattern that was successfully studied
1.1.1.3 ! misho 1507: with a JIT option, the way that the matching is executed is entirely different.
! 1508: However, there is still the possibility of runaway matching that goes on for a
! 1509: very long time, and so the \fImatch_limit\fP value is also used in this case
! 1510: (but in a different way) to limit how long the matching can continue.
1.1 misho 1511: .P
1512: The default value for the limit can be set when PCRE is built; the default
1513: default is 10 million, which handles all but the most extreme cases. You can
1514: override the default by suppling \fBpcre_exec()\fP with a \fBpcre_extra\fP
1515: block in which \fImatch_limit\fP is set, and PCRE_EXTRA_MATCH_LIMIT is set in
1516: the \fIflags\fP field. If the limit is exceeded, \fBpcre_exec()\fP returns
1517: PCRE_ERROR_MATCHLIMIT.
1518: .P
1519: The \fImatch_limit_recursion\fP field is similar to \fImatch_limit\fP, but
1520: instead of limiting the total number of times that \fBmatch()\fP is called, it
1521: limits the depth of recursion. The recursion depth is a smaller number than the
1522: total number of calls, because not all calls to \fBmatch()\fP are recursive.
1523: This limit is of use only if it is set smaller than \fImatch_limit\fP.
1524: .P
1525: Limiting the recursion depth limits the amount of machine stack that can be
1526: used, or, when PCRE has been compiled to use memory on the heap instead of the
1527: stack, the amount of heap memory that can be used. This limit is not relevant,
1.1.1.3 ! misho 1528: and is ignored, when matching is done using JIT compiled code.
1.1 misho 1529: .P
1530: The default value for \fImatch_limit_recursion\fP can be set when PCRE is
1531: built; the default default is the same value as the default for
1532: \fImatch_limit\fP. You can override the default by suppling \fBpcre_exec()\fP
1533: with a \fBpcre_extra\fP block in which \fImatch_limit_recursion\fP is set, and
1534: PCRE_EXTRA_MATCH_LIMIT_RECURSION is set in the \fIflags\fP field. If the limit
1535: is exceeded, \fBpcre_exec()\fP returns PCRE_ERROR_RECURSIONLIMIT.
1536: .P
1537: The \fIcallout_data\fP field is used in conjunction with the "callout" feature,
1538: and is described in the
1539: .\" HREF
1540: \fBpcrecallout\fP
1541: .\"
1542: documentation.
1543: .P
1544: The \fItables\fP field is used to pass a character tables pointer to
1545: \fBpcre_exec()\fP; this overrides the value that is stored with the compiled
1546: pattern. A non-NULL value is stored with the compiled pattern only if custom
1547: tables were supplied to \fBpcre_compile()\fP via its \fItableptr\fP argument.
1548: If NULL is passed to \fBpcre_exec()\fP using this mechanism, it forces PCRE's
1549: internal tables to be used. This facility is helpful when re-using patterns
1550: that have been saved after compiling with an external set of tables, because
1551: the external tables might be at a different address when \fBpcre_exec()\fP is
1552: called. See the
1553: .\" HREF
1554: \fBpcreprecompile\fP
1555: .\"
1556: documentation for a discussion of saving compiled patterns for later use.
1557: .P
1558: If PCRE_EXTRA_MARK is set in the \fIflags\fP field, the \fImark\fP field must
1.1.1.2 misho 1559: be set to point to a suitable variable. If the pattern contains any
1.1 misho 1560: backtracking control verbs such as (*MARK:NAME), and the execution ends up with
1561: a name to pass back, a pointer to the name string (zero terminated) is placed
1562: in the variable pointed to by the \fImark\fP field. The names are within the
1563: compiled pattern; if you wish to retain such a name you must copy it before
1564: freeing the memory of a compiled pattern. If there is no name to pass back, the
1.1.1.2 misho 1565: variable pointed to by the \fImark\fP field is set to NULL. For details of the
1.1 misho 1566: backtracking control verbs, see the section entitled
1567: .\" HTML <a href="pcrepattern#backtrackcontrol">
1568: .\" </a>
1569: "Backtracking control"
1570: .\"
1571: in the
1572: .\" HREF
1573: \fBpcrepattern\fP
1574: .\"
1575: documentation.
1576: .
1577: .
1578: .\" HTML <a name="execoptions"></a>
1579: .SS "Option bits for \fBpcre_exec()\fP"
1580: .rs
1581: .sp
1582: The unused bits of the \fIoptions\fP argument for \fBpcre_exec()\fP must be
1583: zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_\fIxxx\fP,
1584: PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NOTEMPTY_ATSTART,
1.1.1.3 ! misho 1585: PCRE_NO_START_OPTIMIZE, PCRE_NO_UTF8_CHECK, PCRE_PARTIAL_HARD, and
! 1586: PCRE_PARTIAL_SOFT.
1.1 misho 1587: .P
1.1.1.3 ! misho 1588: If the pattern was successfully studied with one of the just-in-time (JIT)
! 1589: compile options, the only supported options for JIT execution are
! 1590: PCRE_NO_UTF8_CHECK, PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY,
! 1591: PCRE_NOTEMPTY_ATSTART, PCRE_PARTIAL_HARD, and PCRE_PARTIAL_SOFT. If an
! 1592: unsupported option is used, JIT execution is disabled and the normal
! 1593: interpretive code in \fBpcre_exec()\fP is run.
1.1 misho 1594: .sp
1595: PCRE_ANCHORED
1596: .sp
1597: The PCRE_ANCHORED option limits \fBpcre_exec()\fP to matching at the first
1598: matching position. If a pattern was compiled with PCRE_ANCHORED, or turned out
1599: to be anchored by virtue of its contents, it cannot be made unachored at
1600: matching time.
1601: .sp
1602: PCRE_BSR_ANYCRLF
1603: PCRE_BSR_UNICODE
1604: .sp
1605: These options (which are mutually exclusive) control what the \eR escape
1606: sequence matches. The choice is either to match only CR, LF, or CRLF, or to
1607: match any Unicode newline sequence. These options override the choice that was
1608: made or defaulted when the pattern was compiled.
1609: .sp
1610: PCRE_NEWLINE_CR
1611: PCRE_NEWLINE_LF
1612: PCRE_NEWLINE_CRLF
1613: PCRE_NEWLINE_ANYCRLF
1614: PCRE_NEWLINE_ANY
1615: .sp
1616: These options override the newline definition that was chosen or defaulted when
1617: the pattern was compiled. For details, see the description of
1618: \fBpcre_compile()\fP above. During matching, the newline choice affects the
1619: behaviour of the dot, circumflex, and dollar metacharacters. It may also alter
1620: the way the match position is advanced after a match failure for an unanchored
1621: pattern.
1622: .P
1623: When PCRE_NEWLINE_CRLF, PCRE_NEWLINE_ANYCRLF, or PCRE_NEWLINE_ANY is set, and a
1624: match attempt for an unanchored pattern fails when the current position is at a
1625: CRLF sequence, and the pattern contains no explicit matches for CR or LF
1626: characters, the match position is advanced by two characters instead of one, in
1627: other words, to after the CRLF.
1628: .P
1629: The above rule is a compromise that makes the most common cases work as
1630: expected. For example, if the pattern is .+A (and the PCRE_DOTALL option is not
1631: set), it does not match the string "\er\enA" because, after failing at the
1632: start, it skips both the CR and the LF before retrying. However, the pattern
1633: [\er\en]A does match that string, because it contains an explicit CR or LF
1634: reference, and so advances only by one character after the first failure.
1635: .P
1636: An explicit match for CR of LF is either a literal appearance of one of those
1637: characters, or one of the \er or \en escape sequences. Implicit matches such as
1638: [^X] do not count, nor does \es (which includes CR and LF in the characters
1639: that it matches).
1640: .P
1641: Notwithstanding the above, anomalous effects may still occur when CRLF is a
1642: valid newline sequence and explicit \er or \en escapes appear in the pattern.
1643: .sp
1644: PCRE_NOTBOL
1645: .sp
1646: This option specifies that first character of the subject string is not the
1647: beginning of a line, so the circumflex metacharacter should not match before
1648: it. Setting this without PCRE_MULTILINE (at compile time) causes circumflex
1649: never to match. This option affects only the behaviour of the circumflex
1650: metacharacter. It does not affect \eA.
1651: .sp
1652: PCRE_NOTEOL
1653: .sp
1654: This option specifies that the end of the subject string is not the end of a
1655: line, so the dollar metacharacter should not match it nor (except in multiline
1656: mode) a newline immediately before it. Setting this without PCRE_MULTILINE (at
1657: compile time) causes dollar never to match. This option affects only the
1658: behaviour of the dollar metacharacter. It does not affect \eZ or \ez.
1659: .sp
1660: PCRE_NOTEMPTY
1661: .sp
1662: An empty string is not considered to be a valid match if this option is set. If
1663: there are alternatives in the pattern, they are tried. If all the alternatives
1664: match the empty string, the entire match fails. For example, if the pattern
1665: .sp
1666: a?b?
1667: .sp
1668: is applied to a string not beginning with "a" or "b", it matches an empty
1669: string at the start of the subject. With PCRE_NOTEMPTY set, this match is not
1670: valid, so PCRE searches further into the string for occurrences of "a" or "b".
1671: .sp
1672: PCRE_NOTEMPTY_ATSTART
1673: .sp
1674: This is like PCRE_NOTEMPTY, except that an empty string match that is not at
1675: the start of the subject is permitted. If the pattern is anchored, such a match
1676: can occur only if the pattern contains \eK.
1677: .P
1678: Perl has no direct equivalent of PCRE_NOTEMPTY or PCRE_NOTEMPTY_ATSTART, but it
1679: does make a special case of a pattern match of the empty string within its
1680: \fBsplit()\fP function, and when using the /g modifier. It is possible to
1681: emulate Perl's behaviour after matching a null string by first trying the match
1682: again at the same offset with PCRE_NOTEMPTY_ATSTART and PCRE_ANCHORED, and then
1683: if that fails, by advancing the starting offset (see below) and trying an
1684: ordinary match again. There is some code that demonstrates how to do this in
1685: the
1686: .\" HREF
1687: \fBpcredemo\fP
1688: .\"
1689: sample program. In the most general case, you have to check to see if the
1690: newline convention recognizes CRLF as a newline, and if so, and the current
1691: character is CR followed by LF, advance the starting offset by two characters
1692: instead of one.
1693: .sp
1694: PCRE_NO_START_OPTIMIZE
1695: .sp
1696: There are a number of optimizations that \fBpcre_exec()\fP uses at the start of
1697: a match, in order to speed up the process. For example, if it is known that an
1698: unanchored match must start with a specific character, it searches the subject
1699: for that character, and fails immediately if it cannot find it, without
1700: actually running the main matching function. This means that a special item
1701: such as (*COMMIT) at the start of a pattern is not considered until after a
1702: suitable starting point for the match has been found. When callouts or (*MARK)
1703: items are in use, these "start-up" optimizations can cause them to be skipped
1704: if the pattern is never actually used. The start-up optimizations are in effect
1705: a pre-scan of the subject that takes place before the pattern is run.
1706: .P
1707: The PCRE_NO_START_OPTIMIZE option disables the start-up optimizations, possibly
1708: causing performance to suffer, but ensuring that in cases where the result is
1709: "no match", the callouts do occur, and that items such as (*COMMIT) and (*MARK)
1710: are considered at every possible starting position in the subject string. If
1711: PCRE_NO_START_OPTIMIZE is set at compile time, it cannot be unset at matching
1.1.1.3 ! misho 1712: time. The use of PCRE_NO_START_OPTIMIZE disables JIT execution; when it is set,
! 1713: matching is always done using interpretively.
1.1 misho 1714: .P
1715: Setting PCRE_NO_START_OPTIMIZE can change the outcome of a matching operation.
1716: Consider the pattern
1717: .sp
1718: (*COMMIT)ABC
1719: .sp
1720: When this is compiled, PCRE records the fact that a match must start with the
1721: character "A". Suppose the subject string is "DEFABC". The start-up
1722: optimization scans along the subject, finds "A" and runs the first match
1723: attempt from there. The (*COMMIT) item means that the pattern must match the
1724: current starting position, which in this case, it does. However, if the same
1725: match is run with PCRE_NO_START_OPTIMIZE set, the initial scan along the
1726: subject string does not happen. The first match attempt is run starting from
1727: "D" and when this fails, (*COMMIT) prevents any further matches being tried, so
1728: the overall result is "no match". If the pattern is studied, more start-up
1729: optimizations may be used. For example, a minimum length for the subject may be
1730: recorded. Consider the pattern
1731: .sp
1732: (*MARK:A)(X|Y)
1733: .sp
1734: The minimum length for a match is one character. If the subject is "ABC", there
1735: will be attempts to match "ABC", "BC", "C", and then finally an empty string.
1736: If the pattern is studied, the final attempt does not take place, because PCRE
1737: knows that the subject is too short, and so the (*MARK) is never encountered.
1738: In this case, studying the pattern does not affect the overall match result,
1739: which is still "no match", but it does affect the auxiliary information that is
1740: returned.
1741: .sp
1742: PCRE_NO_UTF8_CHECK
1743: .sp
1744: When PCRE_UTF8 is set at compile time, the validity of the subject as a UTF-8
1745: string is automatically checked when \fBpcre_exec()\fP is subsequently called.
1.1.1.3 ! misho 1746: The entire string is checked before any other processing takes place. The value
! 1747: of \fIstartoffset\fP is also checked to ensure that it points to the start of a
! 1748: UTF-8 character. There is a discussion about the
! 1749: .\" HTML <a href="pcreunicode.html#utf8strings">
! 1750: .\" </a>
! 1751: validity of UTF-8 strings
! 1752: .\"
! 1753: in the
1.1 misho 1754: .\" HREF
1.1.1.2 misho 1755: \fBpcreunicode\fP
1.1 misho 1756: .\"
1.1.1.2 misho 1757: page. If an invalid sequence of bytes is found, \fBpcre_exec()\fP returns the
1758: error PCRE_ERROR_BADUTF8 or, if PCRE_PARTIAL_HARD is set and the problem is a
1759: truncated character at the end of the subject, PCRE_ERROR_SHORTUTF8. In both
1760: cases, information about the precise nature of the error may also be returned
1761: (see the descriptions of these errors in the section entitled \fIError return
1762: values from\fP \fBpcre_exec()\fP
1.1 misho 1763: .\" HTML <a href="#errorlist">
1764: .\" </a>
1765: below).
1766: .\"
1767: If \fIstartoffset\fP contains a value that does not point to the start of a
1768: UTF-8 character (or to the end of the subject), PCRE_ERROR_BADUTF8_OFFSET is
1769: returned.
1770: .P
1771: If you already know that your subject is valid, and you want to skip these
1772: checks for performance reasons, you can set the PCRE_NO_UTF8_CHECK option when
1773: calling \fBpcre_exec()\fP. You might want to do this for the second and
1774: subsequent calls to \fBpcre_exec()\fP if you are making repeated calls to find
1775: all the matches in a single subject string. However, you should be sure that
1.1.1.2 misho 1776: the value of \fIstartoffset\fP points to the start of a character (or the end
1777: of the subject). When PCRE_NO_UTF8_CHECK is set, the effect of passing an
1778: invalid string as a subject or an invalid value of \fIstartoffset\fP is
1.1 misho 1779: undefined. Your program may crash.
1780: .sp
1781: PCRE_PARTIAL_HARD
1782: PCRE_PARTIAL_SOFT
1783: .sp
1784: These options turn on the partial matching feature. For backwards
1785: compatibility, PCRE_PARTIAL is a synonym for PCRE_PARTIAL_SOFT. A partial match
1786: occurs if the end of the subject string is reached successfully, but there are
1787: not enough subject characters to complete the match. If this happens when
1788: PCRE_PARTIAL_SOFT (but not PCRE_PARTIAL_HARD) is set, matching continues by
1789: testing any remaining alternatives. Only if no complete match can be found is
1790: PCRE_ERROR_PARTIAL returned instead of PCRE_ERROR_NOMATCH. In other words,
1791: PCRE_PARTIAL_SOFT says that the caller is prepared to handle a partial match,
1792: but only if no complete match can be found.
1793: .P
1794: If PCRE_PARTIAL_HARD is set, it overrides PCRE_PARTIAL_SOFT. In this case, if a
1795: partial match is found, \fBpcre_exec()\fP immediately returns
1796: PCRE_ERROR_PARTIAL, without considering any other alternatives. In other words,
1797: when PCRE_PARTIAL_HARD is set, a partial match is considered to be more
1798: important that an alternative complete match.
1799: .P
1800: In both cases, the portion of the string that was inspected when the partial
1801: match was found is set as the first matching string. There is a more detailed
1802: discussion of partial and multi-segment matching, with examples, in the
1803: .\" HREF
1804: \fBpcrepartial\fP
1805: .\"
1806: documentation.
1807: .
1808: .
1809: .SS "The string to be matched by \fBpcre_exec()\fP"
1810: .rs
1811: .sp
1812: The subject string is passed to \fBpcre_exec()\fP as a pointer in
1.1.1.2 misho 1813: \fIsubject\fP, a length in bytes in \fIlength\fP, and a starting byte offset
1.1 misho 1814: in \fIstartoffset\fP. If this is negative or greater than the length of the
1815: subject, \fBpcre_exec()\fP returns PCRE_ERROR_BADOFFSET. When the starting
1816: offset is zero, the search for a match starts at the beginning of the subject,
1817: and this is by far the most common case. In UTF-8 mode, the byte offset must
1818: point to the start of a UTF-8 character (or the end of the subject). Unlike the
1819: pattern string, the subject may contain binary zero bytes.
1820: .P
1821: A non-zero starting offset is useful when searching for another match in the
1822: same subject by calling \fBpcre_exec()\fP again after a previous success.
1823: Setting \fIstartoffset\fP differs from just passing over a shortened string and
1824: setting PCRE_NOTBOL in the case of a pattern that begins with any kind of
1825: lookbehind. For example, consider the pattern
1826: .sp
1827: \eBiss\eB
1828: .sp
1829: which finds occurrences of "iss" in the middle of words. (\eB matches only if
1830: the current position in the subject is not a word boundary.) When applied to
1831: the string "Mississipi" the first call to \fBpcre_exec()\fP finds the first
1832: occurrence. If \fBpcre_exec()\fP is called again with just the remainder of the
1833: subject, namely "issipi", it does not match, because \eB is always false at the
1834: start of the subject, which is deemed to be a word boundary. However, if
1835: \fBpcre_exec()\fP is passed the entire string again, but with \fIstartoffset\fP
1836: set to 4, it finds the second occurrence of "iss" because it is able to look
1837: behind the starting point to discover that it is preceded by a letter.
1838: .P
1839: Finding all the matches in a subject is tricky when the pattern can match an
1840: empty string. It is possible to emulate Perl's /g behaviour by first trying the
1841: match again at the same offset, with the PCRE_NOTEMPTY_ATSTART and
1842: PCRE_ANCHORED options, and then if that fails, advancing the starting offset
1843: and trying an ordinary match again. There is some code that demonstrates how to
1844: do this in the
1845: .\" HREF
1846: \fBpcredemo\fP
1847: .\"
1848: sample program. In the most general case, you have to check to see if the
1849: newline convention recognizes CRLF as a newline, and if so, and the current
1850: character is CR followed by LF, advance the starting offset by two characters
1851: instead of one.
1852: .P
1853: If a non-zero starting offset is passed when the pattern is anchored, one
1854: attempt to match at the given offset is made. This can only succeed if the
1855: pattern does not require the match to be at the start of the subject.
1856: .
1857: .
1858: .SS "How \fBpcre_exec()\fP returns captured substrings"
1859: .rs
1860: .sp
1861: In general, a pattern matches a certain portion of the subject, and in
1862: addition, further substrings from the subject may be picked out by parts of the
1863: pattern. Following the usage in Jeffrey Friedl's book, this is called
1864: "capturing" in what follows, and the phrase "capturing subpattern" is used for
1865: a fragment of a pattern that picks out a substring. PCRE supports several other
1866: kinds of parenthesized subpattern that do not cause substrings to be captured.
1867: .P
1868: Captured substrings are returned to the caller via a vector of integers whose
1869: address is passed in \fIovector\fP. The number of elements in the vector is
1870: passed in \fIovecsize\fP, which must be a non-negative number. \fBNote\fP: this
1871: argument is NOT the size of \fIovector\fP in bytes.
1872: .P
1873: The first two-thirds of the vector is used to pass back captured substrings,
1874: each substring using a pair of integers. The remaining third of the vector is
1875: used as workspace by \fBpcre_exec()\fP while matching capturing subpatterns,
1876: and is not available for passing back information. The number passed in
1877: \fIovecsize\fP should always be a multiple of three. If it is not, it is
1878: rounded down.
1879: .P
1880: When a match is successful, information about captured substrings is returned
1881: in pairs of integers, starting at the beginning of \fIovector\fP, and
1882: continuing up to two-thirds of its length at the most. The first element of
1883: each pair is set to the byte offset of the first character in a substring, and
1884: the second is set to the byte offset of the first character after the end of a
1885: substring. \fBNote\fP: these values are always byte offsets, even in UTF-8
1886: mode. They are not character counts.
1887: .P
1888: The first pair of integers, \fIovector[0]\fP and \fIovector[1]\fP, identify the
1889: portion of the subject string matched by the entire pattern. The next pair is
1890: used for the first capturing subpattern, and so on. The value returned by
1891: \fBpcre_exec()\fP is one more than the highest numbered pair that has been set.
1892: For example, if two substrings have been captured, the returned value is 3. If
1893: there are no capturing subpatterns, the return value from a successful match is
1894: 1, indicating that just the first pair of offsets has been set.
1895: .P
1896: If a capturing subpattern is matched repeatedly, it is the last portion of the
1897: string that it matched that is returned.
1898: .P
1899: If the vector is too small to hold all the captured substring offsets, it is
1900: used as far as possible (up to two-thirds of its length), and the function
1.1.1.3 ! misho 1901: returns a value of zero. If neither the actual string matched nor any captured
1.1 misho 1902: substrings are of interest, \fBpcre_exec()\fP may be called with \fIovector\fP
1903: passed as NULL and \fIovecsize\fP as zero. However, if the pattern contains
1904: back references and the \fIovector\fP is not big enough to remember the related
1905: substrings, PCRE has to get additional memory for use during matching. Thus it
1906: is usually advisable to supply an \fIovector\fP of reasonable size.
1907: .P
1908: There are some cases where zero is returned (indicating vector overflow) when
1909: in fact the vector is exactly the right size for the final match. For example,
1910: consider the pattern
1911: .sp
1912: (a)(?:(b)c|bd)
1913: .sp
1914: If a vector of 6 elements (allowing for only 1 captured substring) is given
1915: with subject string "abd", \fBpcre_exec()\fP will try to set the second
1916: captured string, thereby recording a vector overflow, before failing to match
1917: "c" and backing up to try the second alternative. The zero return, however,
1918: does correctly indicate that the maximum number of slots (namely 2) have been
1919: filled. In similar cases where there is temporary overflow, but the final
1920: number of used slots is actually less than the maximum, a non-zero value is
1921: returned.
1922: .P
1923: The \fBpcre_fullinfo()\fP function can be used to find out how many capturing
1924: subpatterns there are in a compiled pattern. The smallest size for
1925: \fIovector\fP that will allow for \fIn\fP captured substrings, in addition to
1926: the offsets of the substring matched by the whole pattern, is (\fIn\fP+1)*3.
1927: .P
1928: It is possible for capturing subpattern number \fIn+1\fP to match some part of
1929: the subject when subpattern \fIn\fP has not been used at all. For example, if
1930: the string "abc" is matched against the pattern (a|(z))(bc) the return from the
1931: function is 4, and subpatterns 1 and 3 are matched, but 2 is not. When this
1932: happens, both values in the offset pairs corresponding to unused subpatterns
1933: are set to -1.
1934: .P
1935: Offset values that correspond to unused subpatterns at the end of the
1936: expression are also set to -1. For example, if the string "abc" is matched
1937: against the pattern (abc)(x(yz)?)? subpatterns 2 and 3 are not matched. The
1938: return from the function is 2, because the highest used capturing subpattern
1939: number is 1, and the offsets for for the second and third capturing subpatterns
1940: (assuming the vector is large enough, of course) are set to -1.
1941: .P
1942: \fBNote\fP: Elements in the first two-thirds of \fIovector\fP that do not
1943: correspond to capturing parentheses in the pattern are never changed. That is,
1944: if a pattern contains \fIn\fP capturing parentheses, no more than
1945: \fIovector[0]\fP to \fIovector[2n+1]\fP are set by \fBpcre_exec()\fP. The other
1946: elements (in the first two-thirds) retain whatever values they previously had.
1947: .P
1948: Some convenience functions are provided for extracting the captured substrings
1949: as separate strings. These are described below.
1950: .
1951: .
1952: .\" HTML <a name="errorlist"></a>
1953: .SS "Error return values from \fBpcre_exec()\fP"
1954: .rs
1955: .sp
1956: If \fBpcre_exec()\fP fails, it returns a negative number. The following are
1957: defined in the header file:
1958: .sp
1959: PCRE_ERROR_NOMATCH (-1)
1960: .sp
1961: The subject string did not match the pattern.
1962: .sp
1963: PCRE_ERROR_NULL (-2)
1964: .sp
1965: Either \fIcode\fP or \fIsubject\fP was passed as NULL, or \fIovector\fP was
1966: NULL and \fIovecsize\fP was not zero.
1967: .sp
1968: PCRE_ERROR_BADOPTION (-3)
1969: .sp
1970: An unrecognized bit was set in the \fIoptions\fP argument.
1971: .sp
1972: PCRE_ERROR_BADMAGIC (-4)
1973: .sp
1974: PCRE stores a 4-byte "magic number" at the start of the compiled code, to catch
1975: the case when it is passed a junk pointer and to detect when a pattern that was
1976: compiled in an environment of one endianness is run in an environment with the
1977: other endianness. This is the error that PCRE gives when the magic number is
1978: not present.
1979: .sp
1980: PCRE_ERROR_UNKNOWN_OPCODE (-5)
1981: .sp
1982: While running the pattern match, an unknown item was encountered in the
1983: compiled pattern. This error could be caused by a bug in PCRE or by overwriting
1984: of the compiled pattern.
1985: .sp
1986: PCRE_ERROR_NOMEMORY (-6)
1987: .sp
1988: If a pattern contains back references, but the \fIovector\fP that is passed to
1989: \fBpcre_exec()\fP is not big enough to remember the referenced substrings, PCRE
1990: gets a block of memory at the start of matching to use for this purpose. If the
1991: call via \fBpcre_malloc()\fP fails, this error is given. The memory is
1992: automatically freed at the end of matching.
1993: .P
1994: This error is also given if \fBpcre_stack_malloc()\fP fails in
1995: \fBpcre_exec()\fP. This can happen only when PCRE has been compiled with
1996: \fB--disable-stack-for-recursion\fP.
1997: .sp
1998: PCRE_ERROR_NOSUBSTRING (-7)
1999: .sp
2000: This error is used by the \fBpcre_copy_substring()\fP,
2001: \fBpcre_get_substring()\fP, and \fBpcre_get_substring_list()\fP functions (see
2002: below). It is never returned by \fBpcre_exec()\fP.
2003: .sp
2004: PCRE_ERROR_MATCHLIMIT (-8)
2005: .sp
2006: The backtracking limit, as specified by the \fImatch_limit\fP field in a
2007: \fBpcre_extra\fP structure (or defaulted) was reached. See the description
2008: above.
2009: .sp
2010: PCRE_ERROR_CALLOUT (-9)
2011: .sp
2012: This error is never generated by \fBpcre_exec()\fP itself. It is provided for
2013: use by callout functions that want to yield a distinctive error code. See the
2014: .\" HREF
2015: \fBpcrecallout\fP
2016: .\"
2017: documentation for details.
2018: .sp
2019: PCRE_ERROR_BADUTF8 (-10)
2020: .sp
2021: A string that contains an invalid UTF-8 byte sequence was passed as a subject,
2022: and the PCRE_NO_UTF8_CHECK option was not set. If the size of the output vector
2023: (\fIovecsize\fP) is at least 2, the byte offset to the start of the the invalid
2024: UTF-8 character is placed in the first element, and a reason code is placed in
2025: the second element. The reason codes are listed in the
2026: .\" HTML <a href="#badutf8reasons">
2027: .\" </a>
2028: following section.
2029: .\"
2030: For backward compatibility, if PCRE_PARTIAL_HARD is set and the problem is a
2031: truncated UTF-8 character at the end of the subject (reason codes 1 to 5),
2032: PCRE_ERROR_SHORTUTF8 is returned instead of PCRE_ERROR_BADUTF8.
2033: .sp
2034: PCRE_ERROR_BADUTF8_OFFSET (-11)
2035: .sp
2036: The UTF-8 byte sequence that was passed as a subject was checked and found to
2037: be valid (the PCRE_NO_UTF8_CHECK option was not set), but the value of
2038: \fIstartoffset\fP did not point to the beginning of a UTF-8 character or the
2039: end of the subject.
2040: .sp
2041: PCRE_ERROR_PARTIAL (-12)
2042: .sp
2043: The subject string did not match, but it did match partially. See the
2044: .\" HREF
2045: \fBpcrepartial\fP
2046: .\"
2047: documentation for details of partial matching.
2048: .sp
2049: PCRE_ERROR_BADPARTIAL (-13)
2050: .sp
2051: This code is no longer in use. It was formerly returned when the PCRE_PARTIAL
2052: option was used with a compiled pattern containing items that were not
2053: supported for partial matching. From release 8.00 onwards, there are no
2054: restrictions on partial matching.
2055: .sp
2056: PCRE_ERROR_INTERNAL (-14)
2057: .sp
2058: An unexpected internal error has occurred. This error could be caused by a bug
2059: in PCRE or by overwriting of the compiled pattern.
2060: .sp
2061: PCRE_ERROR_BADCOUNT (-15)
2062: .sp
2063: This error is given if the value of the \fIovecsize\fP argument is negative.
2064: .sp
2065: PCRE_ERROR_RECURSIONLIMIT (-21)
2066: .sp
2067: The internal recursion limit, as specified by the \fImatch_limit_recursion\fP
2068: field in a \fBpcre_extra\fP structure (or defaulted) was reached. See the
2069: description above.
2070: .sp
2071: PCRE_ERROR_BADNEWLINE (-23)
2072: .sp
2073: An invalid combination of PCRE_NEWLINE_\fIxxx\fP options was given.
2074: .sp
2075: PCRE_ERROR_BADOFFSET (-24)
2076: .sp
2077: The value of \fIstartoffset\fP was negative or greater than the length of the
2078: subject, that is, the value in \fIlength\fP.
2079: .sp
2080: PCRE_ERROR_SHORTUTF8 (-25)
2081: .sp
2082: This error is returned instead of PCRE_ERROR_BADUTF8 when the subject string
2083: ends with a truncated UTF-8 character and the PCRE_PARTIAL_HARD option is set.
2084: Information about the failure is returned as for PCRE_ERROR_BADUTF8. It is in
2085: fact sufficient to detect this case, but this special error code for
2086: PCRE_PARTIAL_HARD precedes the implementation of returned information; it is
2087: retained for backwards compatibility.
2088: .sp
2089: PCRE_ERROR_RECURSELOOP (-26)
2090: .sp
2091: This error is returned when \fBpcre_exec()\fP detects a recursion loop within
2092: the pattern. Specifically, it means that either the whole pattern or a
2093: subpattern has been called recursively for the second time at the same position
2094: in the subject string. Some simple patterns that might do this are detected and
2095: faulted at compile time, but more complicated cases, in particular mutual
2096: recursions between two different subpatterns, cannot be detected until run
2097: time.
2098: .sp
2099: PCRE_ERROR_JIT_STACKLIMIT (-27)
2100: .sp
1.1.1.3 ! misho 2101: This error is returned when a pattern that was successfully studied using a
! 2102: JIT compile option is being matched, but the memory available for the
! 2103: just-in-time processing stack is not large enough. See the
1.1 misho 2104: .\" HREF
2105: \fBpcrejit\fP
2106: .\"
2107: documentation for more details.
1.1.1.2 misho 2108: .sp
1.1.1.3 ! misho 2109: PCRE_ERROR_BADMODE (-28)
1.1.1.2 misho 2110: .sp
2111: This error is given if a pattern that was compiled by the 8-bit library is
2112: passed to a 16-bit library function, or vice versa.
2113: .sp
1.1.1.3 ! misho 2114: PCRE_ERROR_BADENDIANNESS (-29)
1.1.1.2 misho 2115: .sp
2116: This error is given if a pattern that was compiled and saved is reloaded on a
2117: host with different endianness. The utility function
2118: \fBpcre_pattern_to_host_byte_order()\fP can be used to convert such a pattern
2119: so that it runs on the new host.
1.1 misho 2120: .P
1.1.1.3 ! misho 2121: Error numbers -16 to -20, -22, and -30 are not used by \fBpcre_exec()\fP.
1.1 misho 2122: .
2123: .
2124: .\" HTML <a name="badutf8reasons"></a>
2125: .SS "Reason codes for invalid UTF-8 strings"
2126: .rs
2127: .sp
1.1.1.2 misho 2128: This section applies only to the 8-bit library. The corresponding information
2129: for the 16-bit library is given in the
2130: .\" HREF
2131: \fBpcre16\fP
2132: .\"
2133: page.
2134: .P
1.1 misho 2135: When \fBpcre_exec()\fP returns either PCRE_ERROR_BADUTF8 or
2136: PCRE_ERROR_SHORTUTF8, and the size of the output vector (\fIovecsize\fP) is at
2137: least 2, the offset of the start of the invalid UTF-8 character is placed in
2138: the first output vector element (\fIovector[0]\fP) and a reason code is placed
2139: in the second element (\fIovector[1]\fP). The reason codes are given names in
2140: the \fBpcre.h\fP header file:
2141: .sp
2142: PCRE_UTF8_ERR1
2143: PCRE_UTF8_ERR2
2144: PCRE_UTF8_ERR3
2145: PCRE_UTF8_ERR4
2146: PCRE_UTF8_ERR5
2147: .sp
2148: The string ends with a truncated UTF-8 character; the code specifies how many
2149: bytes are missing (1 to 5). Although RFC 3629 restricts UTF-8 characters to be
2150: no longer than 4 bytes, the encoding scheme (originally defined by RFC 2279)
2151: allows for up to 6 bytes, and this is checked first; hence the possibility of
2152: 4 or 5 missing bytes.
2153: .sp
2154: PCRE_UTF8_ERR6
2155: PCRE_UTF8_ERR7
2156: PCRE_UTF8_ERR8
2157: PCRE_UTF8_ERR9
2158: PCRE_UTF8_ERR10
2159: .sp
2160: The two most significant bits of the 2nd, 3rd, 4th, 5th, or 6th byte of the
2161: character do not have the binary value 0b10 (that is, either the most
2162: significant bit is 0, or the next bit is 1).
2163: .sp
2164: PCRE_UTF8_ERR11
2165: PCRE_UTF8_ERR12
2166: .sp
2167: A character that is valid by the RFC 2279 rules is either 5 or 6 bytes long;
2168: these code points are excluded by RFC 3629.
2169: .sp
2170: PCRE_UTF8_ERR13
2171: .sp
2172: A 4-byte character has a value greater than 0x10fff; these code points are
2173: excluded by RFC 3629.
2174: .sp
2175: PCRE_UTF8_ERR14
2176: .sp
2177: A 3-byte character has a value in the range 0xd800 to 0xdfff; this range of
2178: code points are reserved by RFC 3629 for use with UTF-16, and so are excluded
2179: from UTF-8.
2180: .sp
2181: PCRE_UTF8_ERR15
2182: PCRE_UTF8_ERR16
2183: PCRE_UTF8_ERR17
2184: PCRE_UTF8_ERR18
2185: PCRE_UTF8_ERR19
2186: .sp
2187: A 2-, 3-, 4-, 5-, or 6-byte character is "overlong", that is, it codes for a
2188: value that can be represented by fewer bytes, which is invalid. For example,
2189: the two bytes 0xc0, 0xae give the value 0x2e, whose correct coding uses just
2190: one byte.
2191: .sp
2192: PCRE_UTF8_ERR20
2193: .sp
2194: The two most significant bits of the first byte of a character have the binary
2195: value 0b10 (that is, the most significant bit is 1 and the second is 0). Such a
2196: byte can only validly occur as the second or subsequent byte of a multi-byte
2197: character.
2198: .sp
2199: PCRE_UTF8_ERR21
2200: .sp
2201: The first byte of a character has the value 0xfe or 0xff. These values can
2202: never occur in a valid UTF-8 string.
2203: .
2204: .
2205: .SH "EXTRACTING CAPTURED SUBSTRINGS BY NUMBER"
2206: .rs
2207: .sp
2208: .B int pcre_copy_substring(const char *\fIsubject\fP, int *\fIovector\fP,
2209: .ti +5n
2210: .B int \fIstringcount\fP, int \fIstringnumber\fP, char *\fIbuffer\fP,
2211: .ti +5n
2212: .B int \fIbuffersize\fP);
2213: .PP
2214: .B int pcre_get_substring(const char *\fIsubject\fP, int *\fIovector\fP,
2215: .ti +5n
2216: .B int \fIstringcount\fP, int \fIstringnumber\fP,
2217: .ti +5n
2218: .B const char **\fIstringptr\fP);
2219: .PP
2220: .B int pcre_get_substring_list(const char *\fIsubject\fP,
2221: .ti +5n
2222: .B int *\fIovector\fP, int \fIstringcount\fP, "const char ***\fIlistptr\fP);"
2223: .PP
2224: Captured substrings can be accessed directly by using the offsets returned by
2225: \fBpcre_exec()\fP in \fIovector\fP. For convenience, the functions
2226: \fBpcre_copy_substring()\fP, \fBpcre_get_substring()\fP, and
2227: \fBpcre_get_substring_list()\fP are provided for extracting captured substrings
2228: as new, separate, zero-terminated strings. These functions identify substrings
2229: by number. The next section describes functions for extracting named
2230: substrings.
2231: .P
2232: A substring that contains a binary zero is correctly extracted and has a
2233: further zero added on the end, but the result is not, of course, a C string.
2234: However, you can process such a string by referring to the length that is
2235: returned by \fBpcre_copy_substring()\fP and \fBpcre_get_substring()\fP.
2236: Unfortunately, the interface to \fBpcre_get_substring_list()\fP is not adequate
2237: for handling strings containing binary zeros, because the end of the final
2238: string is not independently indicated.
2239: .P
2240: The first three arguments are the same for all three of these functions:
2241: \fIsubject\fP is the subject string that has just been successfully matched,
2242: \fIovector\fP is a pointer to the vector of integer offsets that was passed to
2243: \fBpcre_exec()\fP, and \fIstringcount\fP is the number of substrings that were
2244: captured by the match, including the substring that matched the entire regular
2245: expression. This is the value returned by \fBpcre_exec()\fP if it is greater
2246: than zero. If \fBpcre_exec()\fP returned zero, indicating that it ran out of
2247: space in \fIovector\fP, the value passed as \fIstringcount\fP should be the
2248: number of elements in the vector divided by three.
2249: .P
2250: The functions \fBpcre_copy_substring()\fP and \fBpcre_get_substring()\fP
2251: extract a single substring, whose number is given as \fIstringnumber\fP. A
2252: value of zero extracts the substring that matched the entire pattern, whereas
2253: higher values extract the captured substrings. For \fBpcre_copy_substring()\fP,
2254: the string is placed in \fIbuffer\fP, whose length is given by
2255: \fIbuffersize\fP, while for \fBpcre_get_substring()\fP a new block of memory is
2256: obtained via \fBpcre_malloc\fP, and its address is returned via
2257: \fIstringptr\fP. The yield of the function is the length of the string, not
2258: including the terminating zero, or one of these error codes:
2259: .sp
2260: PCRE_ERROR_NOMEMORY (-6)
2261: .sp
2262: The buffer was too small for \fBpcre_copy_substring()\fP, or the attempt to get
2263: memory failed for \fBpcre_get_substring()\fP.
2264: .sp
2265: PCRE_ERROR_NOSUBSTRING (-7)
2266: .sp
2267: There is no substring whose number is \fIstringnumber\fP.
2268: .P
2269: The \fBpcre_get_substring_list()\fP function extracts all available substrings
2270: and builds a list of pointers to them. All this is done in a single block of
2271: memory that is obtained via \fBpcre_malloc\fP. The address of the memory block
2272: is returned via \fIlistptr\fP, which is also the start of the list of string
2273: pointers. The end of the list is marked by a NULL pointer. The yield of the
2274: function is zero if all went well, or the error code
2275: .sp
2276: PCRE_ERROR_NOMEMORY (-6)
2277: .sp
2278: if the attempt to get the memory block failed.
2279: .P
2280: When any of these functions encounter a substring that is unset, which can
2281: happen when capturing subpattern number \fIn+1\fP matches some part of the
2282: subject, but subpattern \fIn\fP has not been used at all, they return an empty
2283: string. This can be distinguished from a genuine zero-length substring by
2284: inspecting the appropriate offset in \fIovector\fP, which is negative for unset
2285: substrings.
2286: .P
2287: The two convenience functions \fBpcre_free_substring()\fP and
2288: \fBpcre_free_substring_list()\fP can be used to free the memory returned by
2289: a previous call of \fBpcre_get_substring()\fP or
2290: \fBpcre_get_substring_list()\fP, respectively. They do nothing more than call
2291: the function pointed to by \fBpcre_free\fP, which of course could be called
2292: directly from a C program. However, PCRE is used in some situations where it is
2293: linked via a special interface to another programming language that cannot use
2294: \fBpcre_free\fP directly; it is for these cases that the functions are
2295: provided.
2296: .
2297: .
2298: .SH "EXTRACTING CAPTURED SUBSTRINGS BY NAME"
2299: .rs
2300: .sp
2301: .B int pcre_get_stringnumber(const pcre *\fIcode\fP,
2302: .ti +5n
2303: .B const char *\fIname\fP);
2304: .PP
2305: .B int pcre_copy_named_substring(const pcre *\fIcode\fP,
2306: .ti +5n
2307: .B const char *\fIsubject\fP, int *\fIovector\fP,
2308: .ti +5n
2309: .B int \fIstringcount\fP, const char *\fIstringname\fP,
2310: .ti +5n
2311: .B char *\fIbuffer\fP, int \fIbuffersize\fP);
2312: .PP
2313: .B int pcre_get_named_substring(const pcre *\fIcode\fP,
2314: .ti +5n
2315: .B const char *\fIsubject\fP, int *\fIovector\fP,
2316: .ti +5n
2317: .B int \fIstringcount\fP, const char *\fIstringname\fP,
2318: .ti +5n
2319: .B const char **\fIstringptr\fP);
2320: .PP
2321: To extract a substring by name, you first have to find associated number.
2322: For example, for this pattern
2323: .sp
2324: (a+)b(?<xxx>\ed+)...
2325: .sp
2326: the number of the subpattern called "xxx" is 2. If the name is known to be
2327: unique (PCRE_DUPNAMES was not set), you can find the number from the name by
2328: calling \fBpcre_get_stringnumber()\fP. The first argument is the compiled
2329: pattern, and the second is the name. The yield of the function is the
2330: subpattern number, or PCRE_ERROR_NOSUBSTRING (-7) if there is no subpattern of
2331: that name.
2332: .P
2333: Given the number, you can extract the substring directly, or use one of the
2334: functions described in the previous section. For convenience, there are also
2335: two functions that do the whole job.
2336: .P
2337: Most of the arguments of \fBpcre_copy_named_substring()\fP and
2338: \fBpcre_get_named_substring()\fP are the same as those for the similarly named
2339: functions that extract by number. As these are described in the previous
2340: section, they are not re-described here. There are just two differences:
2341: .P
2342: First, instead of a substring number, a substring name is given. Second, there
2343: is an extra argument, given at the start, which is a pointer to the compiled
2344: pattern. This is needed in order to gain access to the name-to-number
2345: translation table.
2346: .P
2347: These functions call \fBpcre_get_stringnumber()\fP, and if it succeeds, they
2348: then call \fBpcre_copy_substring()\fP or \fBpcre_get_substring()\fP, as
2349: appropriate. \fBNOTE:\fP If PCRE_DUPNAMES is set and there are duplicate names,
2350: the behaviour may not be what you want (see the next section).
2351: .P
2352: \fBWarning:\fP If the pattern uses the (?| feature to set up multiple
2353: subpatterns with the same number, as described in the
2354: .\" HTML <a href="pcrepattern.html#dupsubpatternnumber">
2355: .\" </a>
2356: section on duplicate subpattern numbers
2357: .\"
2358: in the
2359: .\" HREF
2360: \fBpcrepattern\fP
2361: .\"
2362: page, you cannot use names to distinguish the different subpatterns, because
2363: names are not included in the compiled code. The matching process uses only
2364: numbers. For this reason, the use of different names for subpatterns of the
2365: same number causes an error at compile time.
2366: .
2367: .
2368: .SH "DUPLICATE SUBPATTERN NAMES"
2369: .rs
2370: .sp
2371: .B int pcre_get_stringtable_entries(const pcre *\fIcode\fP,
2372: .ti +5n
2373: .B const char *\fIname\fP, char **\fIfirst\fP, char **\fIlast\fP);
2374: .PP
2375: When a pattern is compiled with the PCRE_DUPNAMES option, names for subpatterns
2376: are not required to be unique. (Duplicate names are always allowed for
2377: subpatterns with the same number, created by using the (?| feature. Indeed, if
2378: such subpatterns are named, they are required to use the same names.)
2379: .P
2380: Normally, patterns with duplicate names are such that in any one match, only
2381: one of the named subpatterns participates. An example is shown in the
2382: .\" HREF
2383: \fBpcrepattern\fP
2384: .\"
2385: documentation.
2386: .P
2387: When duplicates are present, \fBpcre_copy_named_substring()\fP and
2388: \fBpcre_get_named_substring()\fP return the first substring corresponding to
2389: the given name that is set. If none are set, PCRE_ERROR_NOSUBSTRING (-7) is
2390: returned; no data is returned. The \fBpcre_get_stringnumber()\fP function
2391: returns one of the numbers that are associated with the name, but it is not
2392: defined which it is.
2393: .P
2394: If you want to get full details of all captured substrings for a given name,
2395: you must use the \fBpcre_get_stringtable_entries()\fP function. The first
2396: argument is the compiled pattern, and the second is the name. The third and
2397: fourth are pointers to variables which are updated by the function. After it
2398: has run, they point to the first and last entries in the name-to-number table
2399: for the given name. The function itself returns the length of each entry, or
2400: PCRE_ERROR_NOSUBSTRING (-7) if there are none. The format of the table is
2401: described above in the section entitled \fIInformation about a pattern\fP
2402: .\" HTML <a href="#infoaboutpattern">
2403: .\" </a>
2404: above.
2405: .\"
2406: Given all the relevant entries for the name, you can extract each of their
2407: numbers, and hence the captured data, if any.
2408: .
2409: .
2410: .SH "FINDING ALL POSSIBLE MATCHES"
2411: .rs
2412: .sp
2413: The traditional matching function uses a similar algorithm to Perl, which stops
2414: when it finds the first match, starting at a given point in the subject. If you
2415: want to find all possible matches, or the longest possible match, consider
2416: using the alternative matching function (see below) instead. If you cannot use
2417: the alternative function, but still need to find all possible matches, you
2418: can kludge it up by making use of the callout facility, which is described in
2419: the
2420: .\" HREF
2421: \fBpcrecallout\fP
2422: .\"
2423: documentation.
2424: .P
2425: What you have to do is to insert a callout right at the end of the pattern.
2426: When your callout function is called, extract and save the current matched
2427: substring. Then return 1, which forces \fBpcre_exec()\fP to backtrack and try
2428: other alternatives. Ultimately, when it runs out of matches, \fBpcre_exec()\fP
2429: will yield PCRE_ERROR_NOMATCH.
2430: .
2431: .
1.1.1.2 misho 2432: .SH "OBTAINING AN ESTIMATE OF STACK USAGE"
2433: .rs
2434: .sp
2435: Matching certain patterns using \fBpcre_exec()\fP can use a lot of process
2436: stack, which in certain environments can be rather limited in size. Some users
2437: find it helpful to have an estimate of the amount of stack that is used by
2438: \fBpcre_exec()\fP, to help them set recursion limits, as described in the
2439: .\" HREF
2440: \fBpcrestack\fP
2441: .\"
2442: documentation. The estimate that is output by \fBpcretest\fP when called with
2443: the \fB-m\fP and \fB-C\fP options is obtained by calling \fBpcre_exec\fP with
2444: the values NULL, NULL, NULL, -999, and -999 for its first five arguments.
2445: .P
2446: Normally, if its first argument is NULL, \fBpcre_exec()\fP immediately returns
2447: the negative error code PCRE_ERROR_NULL, but with this special combination of
2448: arguments, it returns instead a negative number whose absolute value is the
2449: approximate stack frame size in bytes. (A negative number is used so that it is
2450: clear that no match has happened.) The value is approximate because in some
2451: cases, recursive calls to \fBpcre_exec()\fP occur when there are one or two
2452: additional variables on the stack.
2453: .P
2454: If PCRE has been compiled to use the heap instead of the stack for recursion,
2455: the value returned is the size of each block that is obtained from the heap.
2456: .
2457: .
1.1 misho 2458: .\" HTML <a name="dfamatch"></a>
2459: .SH "MATCHING A PATTERN: THE ALTERNATIVE FUNCTION"
2460: .rs
2461: .sp
2462: .B int pcre_dfa_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
2463: .ti +5n
2464: .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
2465: .ti +5n
2466: .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP,
2467: .ti +5n
2468: .B int *\fIworkspace\fP, int \fIwscount\fP);
2469: .P
2470: The function \fBpcre_dfa_exec()\fP is called to match a subject string against
2471: a compiled pattern, using a matching algorithm that scans the subject string
2472: just once, and does not backtrack. This has different characteristics to the
2473: normal algorithm, and is not compatible with Perl. Some of the features of PCRE
2474: patterns are not supported. Nevertheless, there are times when this kind of
2475: matching can be useful. For a discussion of the two matching algorithms, and a
2476: list of features that \fBpcre_dfa_exec()\fP does not support, see the
2477: .\" HREF
2478: \fBpcrematching\fP
2479: .\"
2480: documentation.
2481: .P
2482: The arguments for the \fBpcre_dfa_exec()\fP function are the same as for
2483: \fBpcre_exec()\fP, plus two extras. The \fIovector\fP argument is used in a
2484: different way, and this is described below. The other common arguments are used
2485: in the same way as for \fBpcre_exec()\fP, so their description is not repeated
2486: here.
2487: .P
2488: The two additional arguments provide workspace for the function. The workspace
2489: vector should contain at least 20 elements. It is used for keeping track of
2490: multiple paths through the pattern tree. More workspace will be needed for
2491: patterns and subjects where there are a lot of potential matches.
2492: .P
2493: Here is an example of a simple call to \fBpcre_dfa_exec()\fP:
2494: .sp
2495: int rc;
2496: int ovector[10];
2497: int wspace[20];
2498: rc = pcre_dfa_exec(
2499: re, /* result of pcre_compile() */
2500: NULL, /* we didn't study the pattern */
2501: "some string", /* the subject string */
2502: 11, /* the length of the subject string */
2503: 0, /* start at offset 0 in the subject */
2504: 0, /* default options */
2505: ovector, /* vector of integers for substring information */
2506: 10, /* number of elements (NOT size in bytes) */
2507: wspace, /* working space vector */
2508: 20); /* number of elements (NOT size in bytes) */
2509: .
2510: .SS "Option bits for \fBpcre_dfa_exec()\fP"
2511: .rs
2512: .sp
2513: The unused bits of the \fIoptions\fP argument for \fBpcre_dfa_exec()\fP must be
2514: zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_\fIxxx\fP,
2515: PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NOTEMPTY_ATSTART,
2516: PCRE_NO_UTF8_CHECK, PCRE_BSR_ANYCRLF, PCRE_BSR_UNICODE, PCRE_NO_START_OPTIMIZE,
2517: PCRE_PARTIAL_HARD, PCRE_PARTIAL_SOFT, PCRE_DFA_SHORTEST, and PCRE_DFA_RESTART.
2518: All but the last four of these are exactly the same as for \fBpcre_exec()\fP,
2519: so their description is not repeated here.
2520: .sp
2521: PCRE_PARTIAL_HARD
2522: PCRE_PARTIAL_SOFT
2523: .sp
2524: These have the same general effect as they do for \fBpcre_exec()\fP, but the
2525: details are slightly different. When PCRE_PARTIAL_HARD is set for
2526: \fBpcre_dfa_exec()\fP, it returns PCRE_ERROR_PARTIAL if the end of the subject
2527: is reached and there is still at least one matching possibility that requires
2528: additional characters. This happens even if some complete matches have also
2529: been found. When PCRE_PARTIAL_SOFT is set, the return code PCRE_ERROR_NOMATCH
2530: is converted into PCRE_ERROR_PARTIAL if the end of the subject is reached,
2531: there have been no complete matches, but there is still at least one matching
2532: possibility. The portion of the string that was inspected when the longest
2533: partial match was found is set as the first matching string in both cases.
2534: There is a more detailed discussion of partial and multi-segment matching, with
2535: examples, in the
2536: .\" HREF
2537: \fBpcrepartial\fP
2538: .\"
2539: documentation.
2540: .sp
2541: PCRE_DFA_SHORTEST
2542: .sp
2543: Setting the PCRE_DFA_SHORTEST option causes the matching algorithm to stop as
2544: soon as it has found one match. Because of the way the alternative algorithm
2545: works, this is necessarily the shortest possible match at the first possible
2546: matching point in the subject string.
2547: .sp
2548: PCRE_DFA_RESTART
2549: .sp
2550: When \fBpcre_dfa_exec()\fP returns a partial match, it is possible to call it
2551: again, with additional subject characters, and have it continue with the same
2552: match. The PCRE_DFA_RESTART option requests this action; when it is set, the
2553: \fIworkspace\fP and \fIwscount\fP options must reference the same vector as
2554: before because data about the match so far is left in them after a partial
2555: match. There is more discussion of this facility in the
2556: .\" HREF
2557: \fBpcrepartial\fP
2558: .\"
2559: documentation.
2560: .
2561: .
2562: .SS "Successful returns from \fBpcre_dfa_exec()\fP"
2563: .rs
2564: .sp
2565: When \fBpcre_dfa_exec()\fP succeeds, it may have matched more than one
2566: substring in the subject. Note, however, that all the matches from one run of
2567: the function start at the same point in the subject. The shorter matches are
2568: all initial substrings of the longer matches. For example, if the pattern
2569: .sp
2570: <.*>
2571: .sp
2572: is matched against the string
2573: .sp
2574: This is <something> <something else> <something further> no more
2575: .sp
2576: the three matched strings are
2577: .sp
2578: <something>
2579: <something> <something else>
2580: <something> <something else> <something further>
2581: .sp
2582: On success, the yield of the function is a number greater than zero, which is
2583: the number of matched substrings. The substrings themselves are returned in
2584: \fIovector\fP. Each string uses two elements; the first is the offset to the
2585: start, and the second is the offset to the end. In fact, all the strings have
2586: the same start offset. (Space could have been saved by giving this only once,
2587: but it was decided to retain some compatibility with the way \fBpcre_exec()\fP
2588: returns data, even though the meaning of the strings is different.)
2589: .P
2590: The strings are returned in reverse order of length; that is, the longest
2591: matching string is given first. If there were too many matches to fit into
2592: \fIovector\fP, the yield of the function is zero, and the vector is filled with
2593: the longest matches. Unlike \fBpcre_exec()\fP, \fBpcre_dfa_exec()\fP can use
2594: the entire \fIovector\fP for returning matched strings.
2595: .
2596: .
2597: .SS "Error returns from \fBpcre_dfa_exec()\fP"
2598: .rs
2599: .sp
2600: The \fBpcre_dfa_exec()\fP function returns a negative number when it fails.
2601: Many of the errors are the same as for \fBpcre_exec()\fP, and these are
2602: described
2603: .\" HTML <a href="#errorlist">
2604: .\" </a>
2605: above.
2606: .\"
2607: There are in addition the following errors that are specific to
2608: \fBpcre_dfa_exec()\fP:
2609: .sp
2610: PCRE_ERROR_DFA_UITEM (-16)
2611: .sp
2612: This return is given if \fBpcre_dfa_exec()\fP encounters an item in the pattern
2613: that it does not support, for instance, the use of \eC or a back reference.
2614: .sp
2615: PCRE_ERROR_DFA_UCOND (-17)
2616: .sp
2617: This return is given if \fBpcre_dfa_exec()\fP encounters a condition item that
2618: uses a back reference for the condition, or a test for recursion in a specific
2619: group. These are not supported.
2620: .sp
2621: PCRE_ERROR_DFA_UMLIMIT (-18)
2622: .sp
2623: This return is given if \fBpcre_dfa_exec()\fP is called with an \fIextra\fP
2624: block that contains a setting of the \fImatch_limit\fP or
2625: \fImatch_limit_recursion\fP fields. This is not supported (these fields are
2626: meaningless for DFA matching).
2627: .sp
2628: PCRE_ERROR_DFA_WSSIZE (-19)
2629: .sp
2630: This return is given if \fBpcre_dfa_exec()\fP runs out of space in the
2631: \fIworkspace\fP vector.
2632: .sp
2633: PCRE_ERROR_DFA_RECURSE (-20)
2634: .sp
2635: When a recursive subpattern is processed, the matching function calls itself
2636: recursively, using private vectors for \fIovector\fP and \fIworkspace\fP. This
2637: error is given if the output vector is not large enough. This should be
2638: extremely rare, as a vector of size 1000 is used.
1.1.1.3 ! misho 2639: .sp
! 2640: PCRE_ERROR_DFA_BADRESTART (-30)
! 2641: .sp
! 2642: When \fBpcre_dfa_exec()\fP is called with the \fBPCRE_DFA_RESTART\fP option,
! 2643: some plausibility checks are made on the contents of the workspace, which
! 2644: should contain data about the previous partial match. If any of these checks
! 2645: fail, this error is given.
1.1 misho 2646: .
2647: .
2648: .SH "SEE ALSO"
2649: .rs
2650: .sp
1.1.1.2 misho 2651: \fBpcre16\fP(3), \fBpcrebuild\fP(3), \fBpcrecallout\fP(3), \fBpcrecpp(3)\fP(3),
1.1 misho 2652: \fBpcrematching\fP(3), \fBpcrepartial\fP(3), \fBpcreposix\fP(3),
2653: \fBpcreprecompile\fP(3), \fBpcresample\fP(3), \fBpcrestack\fP(3).
2654: .
2655: .
2656: .SH AUTHOR
2657: .rs
2658: .sp
2659: .nf
2660: Philip Hazel
2661: University Computing Service
2662: Cambridge CB2 3QH, England.
2663: .fi
2664: .
2665: .
2666: .SH REVISION
2667: .rs
2668: .sp
2669: .nf
1.1.1.3 ! misho 2670: Last updated: 17 June 2012
1.1.1.2 misho 2671: Copyright (c) 1997-2012 University of Cambridge.
1.1 misho 2672: .fi
FreeBSD-CVSweb <freebsd-cvsweb@FreeBSD.org>
![[BACK]](/icons/cvsweb/back.gif){kind=icon}
Return to pcreapi.3 CVS log ![[TXT]](/icons/cvsweb/text.gif){kind=icon}
![[DIR]](/icons/cvsweb/dir.gif){kind=icon}
Up to [ELWIX - Embedded LightWeight unIX -] / embedaddon / pcre / doc