Annotation of embedaddon/pcre/doc/pcreapi.3, revision 1.1.1.1
1.1 misho 1: .TH PCREAPI 3
2: .SH NAME
3: PCRE - Perl-compatible regular expressions
4: .SH "PCRE NATIVE API BASIC FUNCTIONS"
5: .rs
6: .sp
7: .B #include <pcre.h>
8: .PP
9: .SM
10: .B pcre *pcre_compile(const char *\fIpattern\fP, int \fIoptions\fP,
11: .ti +5n
12: .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
13: .ti +5n
14: .B const unsigned char *\fItableptr\fP);
15: .PP
16: .B pcre *pcre_compile2(const char *\fIpattern\fP, int \fIoptions\fP,
17: .ti +5n
18: .B int *\fIerrorcodeptr\fP,
19: .ti +5n
20: .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
21: .ti +5n
22: .B const unsigned char *\fItableptr\fP);
23: .PP
24: .B pcre_extra *pcre_study(const pcre *\fIcode\fP, int \fIoptions\fP,
25: .ti +5n
26: .B const char **\fIerrptr\fP);
27: .PP
28: .B void pcre_free_study(pcre_extra *\fIextra\fP);
29: .PP
30: .B int pcre_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
31: .ti +5n
32: .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
33: .ti +5n
34: .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP);
35: .
36: .
37: .SH "PCRE NATIVE API AUXILIARY FUNCTIONS"
38: .rs
39: .sp
40: .B pcre_jit_stack *pcre_jit_stack_alloc(int \fIstartsize\fP, int \fImaxsize\fP);
41: .PP
42: .B void pcre_jit_stack_free(pcre_jit_stack *\fIstack\fP);
43: .PP
44: .B void pcre_assign_jit_stack(pcre_extra *\fIextra\fP,
45: .ti +5n
46: .B pcre_jit_callback \fIcallback\fP, void *\fIdata\fP);
47: .PP
48: .B int pcre_dfa_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
49: .ti +5n
50: .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
51: .ti +5n
52: .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP,
53: .ti +5n
54: .B int *\fIworkspace\fP, int \fIwscount\fP);
55: .PP
56: .B int pcre_copy_named_substring(const pcre *\fIcode\fP,
57: .ti +5n
58: .B const char *\fIsubject\fP, int *\fIovector\fP,
59: .ti +5n
60: .B int \fIstringcount\fP, const char *\fIstringname\fP,
61: .ti +5n
62: .B char *\fIbuffer\fP, int \fIbuffersize\fP);
63: .PP
64: .B int pcre_copy_substring(const char *\fIsubject\fP, int *\fIovector\fP,
65: .ti +5n
66: .B int \fIstringcount\fP, int \fIstringnumber\fP, char *\fIbuffer\fP,
67: .ti +5n
68: .B int \fIbuffersize\fP);
69: .PP
70: .B int pcre_get_named_substring(const pcre *\fIcode\fP,
71: .ti +5n
72: .B const char *\fIsubject\fP, int *\fIovector\fP,
73: .ti +5n
74: .B int \fIstringcount\fP, const char *\fIstringname\fP,
75: .ti +5n
76: .B const char **\fIstringptr\fP);
77: .PP
78: .B int pcre_get_stringnumber(const pcre *\fIcode\fP,
79: .ti +5n
80: .B const char *\fIname\fP);
81: .PP
82: .B int pcre_get_stringtable_entries(const pcre *\fIcode\fP,
83: .ti +5n
84: .B const char *\fIname\fP, char **\fIfirst\fP, char **\fIlast\fP);
85: .PP
86: .B int pcre_get_substring(const char *\fIsubject\fP, int *\fIovector\fP,
87: .ti +5n
88: .B int \fIstringcount\fP, int \fIstringnumber\fP,
89: .ti +5n
90: .B const char **\fIstringptr\fP);
91: .PP
92: .B int pcre_get_substring_list(const char *\fIsubject\fP,
93: .ti +5n
94: .B int *\fIovector\fP, int \fIstringcount\fP, "const char ***\fIlistptr\fP);"
95: .PP
96: .B void pcre_free_substring(const char *\fIstringptr\fP);
97: .PP
98: .B void pcre_free_substring_list(const char **\fIstringptr\fP);
99: .PP
100: .B const unsigned char *pcre_maketables(void);
101: .PP
102: .B int pcre_fullinfo(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
103: .ti +5n
104: .B int \fIwhat\fP, void *\fIwhere\fP);
105: .PP
106: .B int pcre_info(const pcre *\fIcode\fP, int *\fIoptptr\fP, int
107: .B *\fIfirstcharptr\fP);
108: .PP
109: .B int pcre_refcount(pcre *\fIcode\fP, int \fIadjust\fP);
110: .PP
111: .B int pcre_config(int \fIwhat\fP, void *\fIwhere\fP);
112: .PP
113: .B char *pcre_version(void);
114: .
115: .
116: .SH "PCRE NATIVE API INDIRECTED FUNCTIONS"
117: .rs
118: .sp
119: .B void *(*pcre_malloc)(size_t);
120: .PP
121: .B void (*pcre_free)(void *);
122: .PP
123: .B void *(*pcre_stack_malloc)(size_t);
124: .PP
125: .B void (*pcre_stack_free)(void *);
126: .PP
127: .B int (*pcre_callout)(pcre_callout_block *);
128: .
129: .
130: .SH "PCRE API OVERVIEW"
131: .rs
132: .sp
133: PCRE has its own native API, which is described in this document. There are
134: also some wrapper functions that correspond to the POSIX regular expression
135: API, but they do not give access to all the functionality. They are described
136: in the
137: .\" HREF
138: \fBpcreposix\fP
139: .\"
140: documentation. Both of these APIs define a set of C function calls. A C++
141: wrapper is also distributed with PCRE. It is documented in the
142: .\" HREF
143: \fBpcrecpp\fP
144: .\"
145: page.
146: .P
147: The native API C function prototypes are defined in the header file
148: \fBpcre.h\fP, and on Unix systems the library itself is called \fBlibpcre\fP.
149: It can normally be accessed by adding \fB-lpcre\fP to the command for linking
150: an application that uses PCRE. The header file defines the macros PCRE_MAJOR
151: and PCRE_MINOR to contain the major and minor release numbers for the library.
152: Applications can use these to include support for different releases of PCRE.
153: .P
154: In a Windows environment, if you want to statically link an application program
155: against a non-dll \fBpcre.a\fP file, you must define PCRE_STATIC before
156: including \fBpcre.h\fP or \fBpcrecpp.h\fP, because otherwise the
157: \fBpcre_malloc()\fP and \fBpcre_free()\fP exported functions will be declared
158: \fB__declspec(dllimport)\fP, with unwanted results.
159: .P
160: The functions \fBpcre_compile()\fP, \fBpcre_compile2()\fP, \fBpcre_study()\fP,
161: and \fBpcre_exec()\fP are used for compiling and matching regular expressions
162: in a Perl-compatible manner. A sample program that demonstrates the simplest
163: way of using them is provided in the file called \fIpcredemo.c\fP in the PCRE
164: source distribution. A listing of this program is given in the
165: .\" HREF
166: \fBpcredemo\fP
167: .\"
168: documentation, and the
169: .\" HREF
170: \fBpcresample\fP
171: .\"
172: documentation describes how to compile and run it.
173: .P
174: Just-in-time compiler support is an optional feature of PCRE that can be built
175: in appropriate hardware environments. It greatly speeds up the matching
176: performance of many patterns. Simple programs can easily request that it be
177: used if available, by setting an option that is ignored when it is not
178: relevant. More complicated programs might need to make use of the functions
179: \fBpcre_jit_stack_alloc()\fP, \fBpcre_jit_stack_free()\fP, and
180: \fBpcre_assign_jit_stack()\fP in order to control the JIT code's memory usage.
181: These functions are discussed in the
182: .\" HREF
183: \fBpcrejit\fP
184: .\"
185: documentation.
186: .P
187: A second matching function, \fBpcre_dfa_exec()\fP, which is not
188: Perl-compatible, is also provided. This uses a different algorithm for the
189: matching. The alternative algorithm finds all possible matches (at a given
190: point in the subject), and scans the subject just once (unless there are
191: lookbehind assertions). However, this algorithm does not return captured
192: substrings. A description of the two matching algorithms and their advantages
193: and disadvantages is given in the
194: .\" HREF
195: \fBpcrematching\fP
196: .\"
197: documentation.
198: .P
199: In addition to the main compiling and matching functions, there are convenience
200: functions for extracting captured substrings from a subject string that is
201: matched by \fBpcre_exec()\fP. They are:
202: .sp
203: \fBpcre_copy_substring()\fP
204: \fBpcre_copy_named_substring()\fP
205: \fBpcre_get_substring()\fP
206: \fBpcre_get_named_substring()\fP
207: \fBpcre_get_substring_list()\fP
208: \fBpcre_get_stringnumber()\fP
209: \fBpcre_get_stringtable_entries()\fP
210: .sp
211: \fBpcre_free_substring()\fP and \fBpcre_free_substring_list()\fP are also
212: provided, to free the memory used for extracted strings.
213: .P
214: The function \fBpcre_maketables()\fP is used to build a set of character tables
215: in the current locale for passing to \fBpcre_compile()\fP, \fBpcre_exec()\fP,
216: or \fBpcre_dfa_exec()\fP. This is an optional facility that is provided for
217: specialist use. Most commonly, no special tables are passed, in which case
218: internal tables that are generated when PCRE is built are used.
219: .P
220: The function \fBpcre_fullinfo()\fP is used to find out information about a
221: compiled pattern; \fBpcre_info()\fP is an obsolete version that returns only
222: some of the available information, but is retained for backwards compatibility.
223: The function \fBpcre_version()\fP returns a pointer to a string containing the
224: version of PCRE and its date of release.
225: .P
226: The function \fBpcre_refcount()\fP maintains a reference count in a data block
227: containing a compiled pattern. This is provided for the benefit of
228: object-oriented applications.
229: .P
230: The global variables \fBpcre_malloc\fP and \fBpcre_free\fP initially contain
231: the entry points of the standard \fBmalloc()\fP and \fBfree()\fP functions,
232: respectively. PCRE calls the memory management functions via these variables,
233: so a calling program can replace them if it wishes to intercept the calls. This
234: should be done before calling any PCRE functions.
235: .P
236: The global variables \fBpcre_stack_malloc\fP and \fBpcre_stack_free\fP are also
237: indirections to memory management functions. These special functions are used
238: only when PCRE is compiled to use the heap for remembering data, instead of
239: recursive function calls, when running the \fBpcre_exec()\fP function. See the
240: .\" HREF
241: \fBpcrebuild\fP
242: .\"
243: documentation for details of how to do this. It is a non-standard way of
244: building PCRE, for use in environments that have limited stacks. Because of the
245: greater use of memory management, it runs more slowly. Separate functions are
246: provided so that special-purpose external code can be used for this case. When
247: used, these functions are always called in a stack-like manner (last obtained,
248: first freed), and always for memory blocks of the same size. There is a
249: discussion about PCRE's stack usage in the
250: .\" HREF
251: \fBpcrestack\fP
252: .\"
253: documentation.
254: .P
255: The global variable \fBpcre_callout\fP initially contains NULL. It can be set
256: by the caller to a "callout" function, which PCRE will then call at specified
257: points during a matching operation. Details are given in the
258: .\" HREF
259: \fBpcrecallout\fP
260: .\"
261: documentation.
262: .
263: .
264: .\" HTML <a name="newlines"></a>
265: .SH NEWLINES
266: .rs
267: .sp
268: PCRE supports five different conventions for indicating line breaks in
269: strings: a single CR (carriage return) character, a single LF (linefeed)
270: character, the two-character sequence CRLF, any of the three preceding, or any
271: Unicode newline sequence. The Unicode newline sequences are the three just
272: mentioned, plus the single characters VT (vertical tab, U+000B), FF (formfeed,
273: U+000C), NEL (next line, U+0085), LS (line separator, U+2028), and PS
274: (paragraph separator, U+2029).
275: .P
276: Each of the first three conventions is used by at least one operating system as
277: its standard newline sequence. When PCRE is built, a default can be specified.
278: The default default is LF, which is the Unix standard. When PCRE is run, the
279: default can be overridden, either when a pattern is compiled, or when it is
280: matched.
281: .P
282: At compile time, the newline convention can be specified by the \fIoptions\fP
283: argument of \fBpcre_compile()\fP, or it can be specified by special text at the
284: start of the pattern itself; this overrides any other settings. See the
285: .\" HREF
286: \fBpcrepattern\fP
287: .\"
288: page for details of the special character sequences.
289: .P
290: In the PCRE documentation the word "newline" is used to mean "the character or
291: pair of characters that indicate a line break". The choice of newline
292: convention affects the handling of the dot, circumflex, and dollar
293: metacharacters, the handling of #-comments in /x mode, and, when CRLF is a
294: recognized line ending sequence, the match position advancement for a
295: non-anchored pattern. There is more detail about this in the
296: .\" HTML <a href="#execoptions">
297: .\" </a>
298: section on \fBpcre_exec()\fP options
299: .\"
300: below.
301: .P
302: The choice of newline convention does not affect the interpretation of
303: the \en or \er escape sequences, nor does it affect what \eR matches, which is
304: controlled in a similar way, but by separate options.
305: .
306: .
307: .SH MULTITHREADING
308: .rs
309: .sp
310: The PCRE functions can be used in multi-threading applications, with the
311: proviso that the memory management functions pointed to by \fBpcre_malloc\fP,
312: \fBpcre_free\fP, \fBpcre_stack_malloc\fP, and \fBpcre_stack_free\fP, and the
313: callout function pointed to by \fBpcre_callout\fP, are shared by all threads.
314: .P
315: The compiled form of a regular expression is not altered during matching, so
316: the same compiled pattern can safely be used by several threads at once.
317: .P
318: If the just-in-time optimization feature is being used, it needs separate
319: memory stack areas for each thread. See the
320: .\" HREF
321: \fBpcrejit\fP
322: .\"
323: documentation for more details.
324: .
325: .
326: .SH "SAVING PRECOMPILED PATTERNS FOR LATER USE"
327: .rs
328: .sp
329: The compiled form of a regular expression can be saved and re-used at a later
330: time, possibly by a different program, and even on a host other than the one on
331: which it was compiled. Details are given in the
332: .\" HREF
333: \fBpcreprecompile\fP
334: .\"
335: documentation. However, compiling a regular expression with one version of PCRE
336: for use with a different version is not guaranteed to work and may cause
337: crashes.
338: .
339: .
340: .SH "CHECKING BUILD-TIME OPTIONS"
341: .rs
342: .sp
343: .B int pcre_config(int \fIwhat\fP, void *\fIwhere\fP);
344: .PP
345: The function \fBpcre_config()\fP makes it possible for a PCRE client to
346: discover which optional features have been compiled into the PCRE library. The
347: .\" HREF
348: \fBpcrebuild\fP
349: .\"
350: documentation has more details about these optional features.
351: .P
352: The first argument for \fBpcre_config()\fP is an integer, specifying which
353: information is required; the second argument is a pointer to a variable into
354: which the information is placed. The following information is available:
355: .sp
356: PCRE_CONFIG_UTF8
357: .sp
358: The output is an integer that is set to one if UTF-8 support is available;
359: otherwise it is set to zero.
360: .sp
361: PCRE_CONFIG_UNICODE_PROPERTIES
362: .sp
363: The output is an integer that is set to one if support for Unicode character
364: properties is available; otherwise it is set to zero.
365: .sp
366: PCRE_CONFIG_JIT
367: .sp
368: The output is an integer that is set to one if support for just-in-time
369: compiling is available; otherwise it is set to zero.
370: .sp
371: PCRE_CONFIG_NEWLINE
372: .sp
373: The output is an integer whose value specifies the default character sequence
374: that is recognized as meaning "newline". The four values that are supported
375: are: 10 for LF, 13 for CR, 3338 for CRLF, -2 for ANYCRLF, and -1 for ANY.
376: Though they are derived from ASCII, the same values are returned in EBCDIC
377: environments. The default should normally correspond to the standard sequence
378: for your operating system.
379: .sp
380: PCRE_CONFIG_BSR
381: .sp
382: The output is an integer whose value indicates what character sequences the \eR
383: escape sequence matches by default. A value of 0 means that \eR matches any
384: Unicode line ending sequence; a value of 1 means that \eR matches only CR, LF,
385: or CRLF. The default can be overridden when a pattern is compiled or matched.
386: .sp
387: PCRE_CONFIG_LINK_SIZE
388: .sp
389: The output is an integer that contains the number of bytes used for internal
390: linkage in compiled regular expressions. The value is 2, 3, or 4. Larger values
391: allow larger regular expressions to be compiled, at the expense of slower
392: matching. The default value of 2 is sufficient for all but the most massive
393: patterns, since it allows the compiled pattern to be up to 64K in size.
394: .sp
395: PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
396: .sp
397: The output is an integer that contains the threshold above which the POSIX
398: interface uses \fBmalloc()\fP for output vectors. Further details are given in
399: the
400: .\" HREF
401: \fBpcreposix\fP
402: .\"
403: documentation.
404: .sp
405: PCRE_CONFIG_MATCH_LIMIT
406: .sp
407: The output is a long integer that gives the default limit for the number of
408: internal matching function calls in a \fBpcre_exec()\fP execution. Further
409: details are given with \fBpcre_exec()\fP below.
410: .sp
411: PCRE_CONFIG_MATCH_LIMIT_RECURSION
412: .sp
413: The output is a long integer that gives the default limit for the depth of
414: recursion when calling the internal matching function in a \fBpcre_exec()\fP
415: execution. Further details are given with \fBpcre_exec()\fP below.
416: .sp
417: PCRE_CONFIG_STACKRECURSE
418: .sp
419: The output is an integer that is set to one if internal recursion when running
420: \fBpcre_exec()\fP is implemented by recursive function calls that use the stack
421: to remember their state. This is the usual way that PCRE is compiled. The
422: output is zero if PCRE was compiled to use blocks of data on the heap instead
423: of recursive function calls. In this case, \fBpcre_stack_malloc\fP and
424: \fBpcre_stack_free\fP are called to manage memory blocks on the heap, thus
425: avoiding the use of the stack.
426: .
427: .
428: .SH "COMPILING A PATTERN"
429: .rs
430: .sp
431: .B pcre *pcre_compile(const char *\fIpattern\fP, int \fIoptions\fP,
432: .ti +5n
433: .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
434: .ti +5n
435: .B const unsigned char *\fItableptr\fP);
436: .sp
437: .B pcre *pcre_compile2(const char *\fIpattern\fP, int \fIoptions\fP,
438: .ti +5n
439: .B int *\fIerrorcodeptr\fP,
440: .ti +5n
441: .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
442: .ti +5n
443: .B const unsigned char *\fItableptr\fP);
444: .P
445: Either of the functions \fBpcre_compile()\fP or \fBpcre_compile2()\fP can be
446: called to compile a pattern into an internal form. The only difference between
447: the two interfaces is that \fBpcre_compile2()\fP has an additional argument,
448: \fIerrorcodeptr\fP, via which a numerical error code can be returned. To avoid
449: too much repetition, we refer just to \fBpcre_compile()\fP below, but the
450: information applies equally to \fBpcre_compile2()\fP.
451: .P
452: The pattern is a C string terminated by a binary zero, and is passed in the
453: \fIpattern\fP argument. A pointer to a single block of memory that is obtained
454: via \fBpcre_malloc\fP is returned. This contains the compiled code and related
455: data. The \fBpcre\fP type is defined for the returned block; this is a typedef
456: for a structure whose contents are not externally defined. It is up to the
457: caller to free the memory (via \fBpcre_free\fP) when it is no longer required.
458: .P
459: Although the compiled code of a PCRE regex is relocatable, that is, it does not
460: depend on memory location, the complete \fBpcre\fP data block is not
461: fully relocatable, because it may contain a copy of the \fItableptr\fP
462: argument, which is an address (see below).
463: .P
464: The \fIoptions\fP argument contains various bit settings that affect the
465: compilation. It should be zero if no options are required. The available
466: options are described below. Some of them (in particular, those that are
467: compatible with Perl, but some others as well) can also be set and unset from
468: within the pattern (see the detailed description in the
469: .\" HREF
470: \fBpcrepattern\fP
471: .\"
472: documentation). For those options that can be different in different parts of
473: the pattern, the contents of the \fIoptions\fP argument specifies their
474: settings at the start of compilation and execution. The PCRE_ANCHORED,
475: PCRE_BSR_\fIxxx\fP, PCRE_NEWLINE_\fIxxx\fP, PCRE_NO_UTF8_CHECK, and
476: PCRE_NO_START_OPT options can be set at the time of matching as well as at
477: compile time.
478: .P
479: If \fIerrptr\fP is NULL, \fBpcre_compile()\fP returns NULL immediately.
480: Otherwise, if compilation of a pattern fails, \fBpcre_compile()\fP returns
481: NULL, and sets the variable pointed to by \fIerrptr\fP to point to a textual
482: error message. This is a static string that is part of the library. You must
483: not try to free it. Normally, the offset from the start of the pattern to the
484: byte that was being processed when the error was discovered is placed in the
485: variable pointed to by \fIerroffset\fP, which must not be NULL (if it is, an
486: immediate error is given). However, for an invalid UTF-8 string, the offset is
487: that of the first byte of the failing character. Also, some errors are not
488: detected until checks are carried out when the whole pattern has been scanned;
489: in these cases the offset passed back is the length of the pattern.
490: .P
491: Note that the offset is in bytes, not characters, even in UTF-8 mode. It may
492: sometimes point into the middle of a UTF-8 character.
493: .P
494: If \fBpcre_compile2()\fP is used instead of \fBpcre_compile()\fP, and the
495: \fIerrorcodeptr\fP argument is not NULL, a non-zero error code number is
496: returned via this argument in the event of an error. This is in addition to the
497: textual error message. Error codes and messages are listed below.
498: .P
499: If the final argument, \fItableptr\fP, is NULL, PCRE uses a default set of
500: character tables that are built when PCRE is compiled, using the default C
501: locale. Otherwise, \fItableptr\fP must be an address that is the result of a
502: call to \fBpcre_maketables()\fP. This value is stored with the compiled
503: pattern, and used again by \fBpcre_exec()\fP, unless another table pointer is
504: passed to it. For more discussion, see the section on locale support below.
505: .P
506: This code fragment shows a typical straightforward call to \fBpcre_compile()\fP:
507: .sp
508: pcre *re;
509: const char *error;
510: int erroffset;
511: re = pcre_compile(
512: "^A.*Z", /* the pattern */
513: 0, /* default options */
514: &error, /* for error message */
515: &erroffset, /* for error offset */
516: NULL); /* use default character tables */
517: .sp
518: The following names for option bits are defined in the \fBpcre.h\fP header
519: file:
520: .sp
521: PCRE_ANCHORED
522: .sp
523: If this bit is set, the pattern is forced to be "anchored", that is, it is
524: constrained to match only at the first matching point in the string that is
525: being searched (the "subject string"). This effect can also be achieved by
526: appropriate constructs in the pattern itself, which is the only way to do it in
527: Perl.
528: .sp
529: PCRE_AUTO_CALLOUT
530: .sp
531: If this bit is set, \fBpcre_compile()\fP automatically inserts callout items,
532: all with number 255, before each pattern item. For discussion of the callout
533: facility, see the
534: .\" HREF
535: \fBpcrecallout\fP
536: .\"
537: documentation.
538: .sp
539: PCRE_BSR_ANYCRLF
540: PCRE_BSR_UNICODE
541: .sp
542: These options (which are mutually exclusive) control what the \eR escape
543: sequence matches. The choice is either to match only CR, LF, or CRLF, or to
544: match any Unicode newline sequence. The default is specified when PCRE is
545: built. It can be overridden from within the pattern, or by setting an option
546: when a compiled pattern is matched.
547: .sp
548: PCRE_CASELESS
549: .sp
550: If this bit is set, letters in the pattern match both upper and lower case
551: letters. It is equivalent to Perl's /i option, and it can be changed within a
552: pattern by a (?i) option setting. In UTF-8 mode, PCRE always understands the
553: concept of case for characters whose values are less than 128, so caseless
554: matching is always possible. For characters with higher values, the concept of
555: case is supported if PCRE is compiled with Unicode property support, but not
556: otherwise. If you want to use caseless matching for characters 128 and above,
557: you must ensure that PCRE is compiled with Unicode property support as well as
558: with UTF-8 support.
559: .sp
560: PCRE_DOLLAR_ENDONLY
561: .sp
562: If this bit is set, a dollar metacharacter in the pattern matches only at the
563: end of the subject string. Without this option, a dollar also matches
564: immediately before a newline at the end of the string (but not before any other
565: newlines). The PCRE_DOLLAR_ENDONLY option is ignored if PCRE_MULTILINE is set.
566: There is no equivalent to this option in Perl, and no way to set it within a
567: pattern.
568: .sp
569: PCRE_DOTALL
570: .sp
571: If this bit is set, a dot metacharacter in the pattern matches a character of
572: any value, including one that indicates a newline. However, it only ever
573: matches one character, even if newlines are coded as CRLF. Without this option,
574: a dot does not match when the current position is at a newline. This option is
575: equivalent to Perl's /s option, and it can be changed within a pattern by a
576: (?s) option setting. A negative class such as [^a] always matches newline
577: characters, independent of the setting of this option.
578: .sp
579: PCRE_DUPNAMES
580: .sp
581: If this bit is set, names used to identify capturing subpatterns need not be
582: unique. This can be helpful for certain types of pattern when it is known that
583: only one instance of the named subpattern can ever be matched. There are more
584: details of named subpatterns below; see also the
585: .\" HREF
586: \fBpcrepattern\fP
587: .\"
588: documentation.
589: .sp
590: PCRE_EXTENDED
591: .sp
592: If this bit is set, whitespace data characters in the pattern are totally
593: ignored except when escaped or inside a character class. Whitespace does not
594: include the VT character (code 11). In addition, characters between an
595: unescaped # outside a character class and the next newline, inclusive, are also
596: ignored. This is equivalent to Perl's /x option, and it can be changed within a
597: pattern by a (?x) option setting.
598: .P
599: Which characters are interpreted as newlines is controlled by the options
600: passed to \fBpcre_compile()\fP or by a special sequence at the start of the
601: pattern, as described in the section entitled
602: .\" HTML <a href="pcrepattern.html#newlines">
603: .\" </a>
604: "Newline conventions"
605: .\"
606: in the \fBpcrepattern\fP documentation. Note that the end of this type of
607: comment is a literal newline sequence in the pattern; escape sequences that
608: happen to represent a newline do not count.
609: .P
610: This option makes it possible to include comments inside complicated patterns.
611: Note, however, that this applies only to data characters. Whitespace characters
612: may never appear within special character sequences in a pattern, for example
613: within the sequence (?( that introduces a conditional subpattern.
614: .sp
615: PCRE_EXTRA
616: .sp
617: This option was invented in order to turn on additional functionality of PCRE
618: that is incompatible with Perl, but it is currently of very little use. When
619: set, any backslash in a pattern that is followed by a letter that has no
620: special meaning causes an error, thus reserving these combinations for future
621: expansion. By default, as in Perl, a backslash followed by a letter with no
622: special meaning is treated as a literal. (Perl can, however, be persuaded to
623: give an error for this, by running it with the -w option.) There are at present
624: no other features controlled by this option. It can also be set by a (?X)
625: option setting within a pattern.
626: .sp
627: PCRE_FIRSTLINE
628: .sp
629: If this option is set, an unanchored pattern is required to match before or at
630: the first newline in the subject string, though the matched text may continue
631: over the newline.
632: .sp
633: PCRE_JAVASCRIPT_COMPAT
634: .sp
635: If this option is set, PCRE's behaviour is changed in some ways so that it is
636: compatible with JavaScript rather than Perl. The changes are as follows:
637: .P
638: (1) A lone closing square bracket in a pattern causes a compile-time error,
639: because this is illegal in JavaScript (by default it is treated as a data
640: character). Thus, the pattern AB]CD becomes illegal when this option is set.
641: .P
642: (2) At run time, a back reference to an unset subpattern group matches an empty
643: string (by default this causes the current matching alternative to fail). A
644: pattern such as (\e1)(a) succeeds when this option is set (assuming it can find
645: an "a" in the subject), whereas it fails by default, for Perl compatibility.
646: .P
647: (3) \eU matches an upper case "U" character; by default \eU causes a compile
648: time error (Perl uses \eU to upper case subsequent characters).
649: .P
650: (4) \eu matches a lower case "u" character unless it is followed by four
651: hexadecimal digits, in which case the hexadecimal number defines the code point
652: to match. By default, \eu causes a compile time error (Perl uses it to upper
653: case the following character).
654: .P
655: (5) \ex matches a lower case "x" character unless it is followed by two
656: hexadecimal digits, in which case the hexadecimal number defines the code point
657: to match. By default, as in Perl, a hexadecimal number is always expected after
658: \ex, but it may have zero, one, or two digits (so, for example, \exz matches a
659: binary zero character followed by z).
660: .sp
661: PCRE_MULTILINE
662: .sp
663: By default, PCRE treats the subject string as consisting of a single line of
664: characters (even if it actually contains newlines). The "start of line"
665: metacharacter (^) matches only at the start of the string, while the "end of
666: line" metacharacter ($) matches only at the end of the string, or before a
667: terminating newline (unless PCRE_DOLLAR_ENDONLY is set). This is the same as
668: Perl.
669: .P
670: When PCRE_MULTILINE it is set, the "start of line" and "end of line" constructs
671: match immediately following or immediately before internal newlines in the
672: subject string, respectively, as well as at the very start and end. This is
673: equivalent to Perl's /m option, and it can be changed within a pattern by a
674: (?m) option setting. If there are no newlines in a subject string, or no
675: occurrences of ^ or $ in a pattern, setting PCRE_MULTILINE has no effect.
676: .sp
677: PCRE_NEWLINE_CR
678: PCRE_NEWLINE_LF
679: PCRE_NEWLINE_CRLF
680: PCRE_NEWLINE_ANYCRLF
681: PCRE_NEWLINE_ANY
682: .sp
683: These options override the default newline definition that was chosen when PCRE
684: was built. Setting the first or the second specifies that a newline is
685: indicated by a single character (CR or LF, respectively). Setting
686: PCRE_NEWLINE_CRLF specifies that a newline is indicated by the two-character
687: CRLF sequence. Setting PCRE_NEWLINE_ANYCRLF specifies that any of the three
688: preceding sequences should be recognized. Setting PCRE_NEWLINE_ANY specifies
689: that any Unicode newline sequence should be recognized. The Unicode newline
690: sequences are the three just mentioned, plus the single characters VT (vertical
691: tab, U+000B), FF (formfeed, U+000C), NEL (next line, U+0085), LS (line
692: separator, U+2028), and PS (paragraph separator, U+2029). The last two are
693: recognized only in UTF-8 mode.
694: .P
695: The newline setting in the options word uses three bits that are treated
696: as a number, giving eight possibilities. Currently only six are used (default
697: plus the five values above). This means that if you set more than one newline
698: option, the combination may or may not be sensible. For example,
699: PCRE_NEWLINE_CR with PCRE_NEWLINE_LF is equivalent to PCRE_NEWLINE_CRLF, but
700: other combinations may yield unused numbers and cause an error.
701: .P
702: The only time that a line break in a pattern is specially recognized when
703: compiling is when PCRE_EXTENDED is set. CR and LF are whitespace characters,
704: and so are ignored in this mode. Also, an unescaped # outside a character class
705: indicates a comment that lasts until after the next line break sequence. In
706: other circumstances, line break sequences in patterns are treated as literal
707: data.
708: .P
709: The newline option that is set at compile time becomes the default that is used
710: for \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP, but it can be overridden.
711: .sp
712: PCRE_NO_AUTO_CAPTURE
713: .sp
714: If this option is set, it disables the use of numbered capturing parentheses in
715: the pattern. Any opening parenthesis that is not followed by ? behaves as if it
716: were followed by ?: but named parentheses can still be used for capturing (and
717: they acquire numbers in the usual way). There is no equivalent of this option
718: in Perl.
719: .sp
720: NO_START_OPTIMIZE
721: .sp
722: This is an option that acts at matching time; that is, it is really an option
723: for \fBpcre_exec()\fP or \fBpcre_dfa_exec()\fP. If it is set at compile time,
724: it is remembered with the compiled pattern and assumed at matching time. For
725: details see the discussion of PCRE_NO_START_OPTIMIZE
726: .\" HTML <a href="#execoptions">
727: .\" </a>
728: below.
729: .\"
730: .sp
731: PCRE_UCP
732: .sp
733: This option changes the way PCRE processes \eB, \eb, \eD, \ed, \eS, \es, \eW,
734: \ew, and some of the POSIX character classes. By default, only ASCII characters
735: are recognized, but if PCRE_UCP is set, Unicode properties are used instead to
736: classify characters. More details are given in the section on
737: .\" HTML <a href="pcre.html#genericchartypes">
738: .\" </a>
739: generic character types
740: .\"
741: in the
742: .\" HREF
743: \fBpcrepattern\fP
744: .\"
745: page. If you set PCRE_UCP, matching one of the items it affects takes much
746: longer. The option is available only if PCRE has been compiled with Unicode
747: property support.
748: .sp
749: PCRE_UNGREEDY
750: .sp
751: This option inverts the "greediness" of the quantifiers so that they are not
752: greedy by default, but become greedy if followed by "?". It is not compatible
753: with Perl. It can also be set by a (?U) option setting within the pattern.
754: .sp
755: PCRE_UTF8
756: .sp
757: This option causes PCRE to regard both the pattern and the subject as strings
758: of UTF-8 characters instead of single-byte character strings. However, it is
759: available only when PCRE is built to include UTF-8 support. If not, the use
760: of this option provokes an error. Details of how this option changes the
761: behaviour of PCRE are given in the
762: .\" HREF
763: \fBpcreunicode\fP
764: .\"
765: page.
766: .sp
767: PCRE_NO_UTF8_CHECK
768: .sp
769: When PCRE_UTF8 is set, the validity of the pattern as a UTF-8 string is
770: automatically checked. There is a discussion about the
771: .\" HTML <a href="pcre.html#utf8strings">
772: .\" </a>
773: validity of UTF-8 strings
774: .\"
775: in the main
776: .\" HREF
777: \fBpcre\fP
778: .\"
779: page. If an invalid UTF-8 sequence of bytes is found, \fBpcre_compile()\fP
780: returns an error. If you already know that your pattern is valid, and you want
781: to skip this check for performance reasons, you can set the PCRE_NO_UTF8_CHECK
782: option. When it is set, the effect of passing an invalid UTF-8 string as a
783: pattern is undefined. It may cause your program to crash. Note that this option
784: can also be passed to \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP, to suppress
785: the UTF-8 validity checking of subject strings.
786: .
787: .
788: .SH "COMPILATION ERROR CODES"
789: .rs
790: .sp
791: The following table lists the error codes than may be returned by
792: \fBpcre_compile2()\fP, along with the error messages that may be returned by
793: both compiling functions. As PCRE has developed, some error codes have fallen
794: out of use. To avoid confusion, they have not been re-used.
795: .sp
796: 0 no error
797: 1 \e at end of pattern
798: 2 \ec at end of pattern
799: 3 unrecognized character follows \e
800: 4 numbers out of order in {} quantifier
801: 5 number too big in {} quantifier
802: 6 missing terminating ] for character class
803: 7 invalid escape sequence in character class
804: 8 range out of order in character class
805: 9 nothing to repeat
806: 10 [this code is not in use]
807: 11 internal error: unexpected repeat
808: 12 unrecognized character after (? or (?-
809: 13 POSIX named classes are supported only within a class
810: 14 missing )
811: 15 reference to non-existent subpattern
812: 16 erroffset passed as NULL
813: 17 unknown option bit(s) set
814: 18 missing ) after comment
815: 19 [this code is not in use]
816: 20 regular expression is too large
817: 21 failed to get memory
818: 22 unmatched parentheses
819: 23 internal error: code overflow
820: 24 unrecognized character after (?<
821: 25 lookbehind assertion is not fixed length
822: 26 malformed number or name after (?(
823: 27 conditional group contains more than two branches
824: 28 assertion expected after (?(
825: 29 (?R or (?[+-]digits must be followed by )
826: 30 unknown POSIX class name
827: 31 POSIX collating elements are not supported
828: 32 this version of PCRE is not compiled with PCRE_UTF8 support
829: 33 [this code is not in use]
830: 34 character value in \ex{...} sequence is too large
831: 35 invalid condition (?(0)
832: 36 \eC not allowed in lookbehind assertion
833: 37 PCRE does not support \eL, \el, \eN{name}, \eU, or \eu
834: 38 number after (?C is > 255
835: 39 closing ) for (?C expected
836: 40 recursive call could loop indefinitely
837: 41 unrecognized character after (?P
838: 42 syntax error in subpattern name (missing terminator)
839: 43 two named subpatterns have the same name
840: 44 invalid UTF-8 string
841: 45 support for \eP, \ep, and \eX has not been compiled
842: 46 malformed \eP or \ep sequence
843: 47 unknown property name after \eP or \ep
844: 48 subpattern name is too long (maximum 32 characters)
845: 49 too many named subpatterns (maximum 10000)
846: 50 [this code is not in use]
847: 51 octal value is greater than \e377 (not in UTF-8 mode)
848: 52 internal error: overran compiling workspace
849: 53 internal error: previously-checked referenced subpattern
850: not found
851: 54 DEFINE group contains more than one branch
852: 55 repeating a DEFINE group is not allowed
853: 56 inconsistent NEWLINE options
854: 57 \eg is not followed by a braced, angle-bracketed, or quoted
855: name/number or by a plain number
856: 58 a numbered reference must not be zero
857: 59 an argument is not allowed for (*ACCEPT), (*FAIL), or (*COMMIT)
858: 60 (*VERB) not recognized
859: 61 number is too big
860: 62 subpattern name expected
861: 63 digit expected after (?+
862: 64 ] is an invalid data character in JavaScript compatibility mode
863: 65 different names for subpatterns of the same number are
864: not allowed
865: 66 (*MARK) must have an argument
866: 67 this version of PCRE is not compiled with PCRE_UCP support
867: 68 \ec must be followed by an ASCII character
868: 69 \ek is not followed by a braced, angle-bracketed, or quoted name
869: .sp
870: The numbers 32 and 10000 in errors 48 and 49 are defaults; different values may
871: be used if the limits were changed when PCRE was built.
872: .
873: .
874: .\" HTML <a name="studyingapattern"></a>
875: .SH "STUDYING A PATTERN"
876: .rs
877: .sp
878: .B pcre_extra *pcre_study(const pcre *\fIcode\fP, int \fIoptions\fP
879: .ti +5n
880: .B const char **\fIerrptr\fP);
881: .PP
882: If a compiled pattern is going to be used several times, it is worth spending
883: more time analyzing it in order to speed up the time taken for matching. The
884: function \fBpcre_study()\fP takes a pointer to a compiled pattern as its first
885: argument. If studying the pattern produces additional information that will
886: help speed up matching, \fBpcre_study()\fP returns a pointer to a
887: \fBpcre_extra\fP block, in which the \fIstudy_data\fP field points to the
888: results of the study.
889: .P
890: The returned value from \fBpcre_study()\fP can be passed directly to
891: \fBpcre_exec()\fP or \fBpcre_dfa_exec()\fP. However, a \fBpcre_extra\fP block
892: also contains other fields that can be set by the caller before the block is
893: passed; these are described
894: .\" HTML <a href="#extradata">
895: .\" </a>
896: below
897: .\"
898: in the section on matching a pattern.
899: .P
900: If studying the pattern does not produce any useful information,
901: \fBpcre_study()\fP returns NULL. In that circumstance, if the calling program
902: wants to pass any of the other fields to \fBpcre_exec()\fP or
903: \fBpcre_dfa_exec()\fP, it must set up its own \fBpcre_extra\fP block.
904: .P
905: The second argument of \fBpcre_study()\fP contains option bits. There is only
906: one option: PCRE_STUDY_JIT_COMPILE. If this is set, and the just-in-time
907: compiler is available, the pattern is further compiled into machine code that
908: executes much faster than the \fBpcre_exec()\fP matching function. If
909: the just-in-time compiler is not available, this option is ignored. All other
910: bits in the \fIoptions\fP argument must be zero.
911: .P
912: JIT compilation is a heavyweight optimization. It can take some time for
913: patterns to be analyzed, and for one-off matches and simple patterns the
914: benefit of faster execution might be offset by a much slower study time.
915: Not all patterns can be optimized by the JIT compiler. For those that cannot be
916: handled, matching automatically falls back to the \fBpcre_exec()\fP
917: interpreter. For more details, see the
918: .\" HREF
919: \fBpcrejit\fP
920: .\"
921: documentation.
922: .P
923: The third argument for \fBpcre_study()\fP is a pointer for an error message. If
924: studying succeeds (even if no data is returned), the variable it points to is
925: set to NULL. Otherwise it is set to point to a textual error message. This is a
926: static string that is part of the library. You must not try to free it. You
927: should test the error pointer for NULL after calling \fBpcre_study()\fP, to be
928: sure that it has run successfully.
929: .P
930: When you are finished with a pattern, you can free the memory used for the
931: study data by calling \fBpcre_free_study()\fP. This function was added to the
932: API for release 8.20. For earlier versions, the memory could be freed with
933: \fBpcre_free()\fP, just like the pattern itself. This will still work in cases
934: where PCRE_STUDY_JIT_COMPILE is not used, but it is advisable to change to the
935: new function when convenient.
936: .P
937: This is a typical way in which \fBpcre_study\fP() is used (except that in a
938: real application there should be tests for errors):
939: .sp
940: int rc;
941: pcre *re;
942: pcre_extra *sd;
943: re = pcre_compile("pattern", 0, &error, &erroroffset, NULL);
944: sd = pcre_study(
945: re, /* result of pcre_compile() */
946: 0, /* no options */
947: &error); /* set to NULL or points to a message */
948: rc = pcre_exec( /* see below for details of pcre_exec() options */
949: re, sd, "subject", 7, 0, 0, ovector, 30);
950: ...
951: pcre_free_study(sd);
952: pcre_free(re);
953: .sp
954: Studying a pattern does two things: first, a lower bound for the length of
955: subject string that is needed to match the pattern is computed. This does not
956: mean that there are any strings of that length that match, but it does
957: guarantee that no shorter strings match. The value is used by
958: \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP to avoid wasting time by trying to
959: match strings that are shorter than the lower bound. You can find out the value
960: in a calling program via the \fBpcre_fullinfo()\fP function.
961: .P
962: Studying a pattern is also useful for non-anchored patterns that do not have a
963: single fixed starting character. A bitmap of possible starting bytes is
964: created. This speeds up finding a position in the subject at which to start
965: matching.
966: .P
967: These two optimizations apply to both \fBpcre_exec()\fP and
968: \fBpcre_dfa_exec()\fP. However, they are not used by \fBpcre_exec()\fP if
969: \fBpcre_study()\fP is called with the PCRE_STUDY_JIT_COMPILE option, and
970: just-in-time compiling is successful. The optimizations can be disabled by
971: setting the PCRE_NO_START_OPTIMIZE option when calling \fBpcre_exec()\fP or
972: \fBpcre_dfa_exec()\fP. You might want to do this if your pattern contains
973: callouts or (*MARK) (which cannot be handled by the JIT compiler), and you want
974: to make use of these facilities in cases where matching fails. See the
975: discussion of PCRE_NO_START_OPTIMIZE
976: .\" HTML <a href="#execoptions">
977: .\" </a>
978: below.
979: .\"
980: .
981: .
982: .\" HTML <a name="localesupport"></a>
983: .SH "LOCALE SUPPORT"
984: .rs
985: .sp
986: PCRE handles caseless matching, and determines whether characters are letters,
987: digits, or whatever, by reference to a set of tables, indexed by character
988: value. When running in UTF-8 mode, this applies only to characters with codes
989: less than 128. By default, higher-valued codes never match escapes such as \ew
990: or \ed, but they can be tested with \ep if PCRE is built with Unicode character
991: property support. Alternatively, the PCRE_UCP option can be set at compile
992: time; this causes \ew and friends to use Unicode property support instead of
993: built-in tables. The use of locales with Unicode is discouraged. If you are
994: handling characters with codes greater than 128, you should either use UTF-8
995: and Unicode, or use locales, but not try to mix the two.
996: .P
997: PCRE contains an internal set of tables that are used when the final argument
998: of \fBpcre_compile()\fP is NULL. These are sufficient for many applications.
999: Normally, the internal tables recognize only ASCII characters. However, when
1000: PCRE is built, it is possible to cause the internal tables to be rebuilt in the
1001: default "C" locale of the local system, which may cause them to be different.
1002: .P
1003: The internal tables can always be overridden by tables supplied by the
1004: application that calls PCRE. These may be created in a different locale from
1005: the default. As more and more applications change to using Unicode, the need
1006: for this locale support is expected to die away.
1007: .P
1008: External tables are built by calling the \fBpcre_maketables()\fP function,
1009: which has no arguments, in the relevant locale. The result can then be passed
1010: to \fBpcre_compile()\fP or \fBpcre_exec()\fP as often as necessary. For
1011: example, to build and use tables that are appropriate for the French locale
1012: (where accented characters with values greater than 128 are treated as letters),
1013: the following code could be used:
1014: .sp
1015: setlocale(LC_CTYPE, "fr_FR");
1016: tables = pcre_maketables();
1017: re = pcre_compile(..., tables);
1018: .sp
1019: The locale name "fr_FR" is used on Linux and other Unix-like systems; if you
1020: are using Windows, the name for the French locale is "french".
1021: .P
1022: When \fBpcre_maketables()\fP runs, the tables are built in memory that is
1023: obtained via \fBpcre_malloc\fP. It is the caller's responsibility to ensure
1024: that the memory containing the tables remains available for as long as it is
1025: needed.
1026: .P
1027: The pointer that is passed to \fBpcre_compile()\fP is saved with the compiled
1028: pattern, and the same tables are used via this pointer by \fBpcre_study()\fP
1029: and normally also by \fBpcre_exec()\fP. Thus, by default, for any single
1030: pattern, compilation, studying and matching all happen in the same locale, but
1031: different patterns can be compiled in different locales.
1032: .P
1033: It is possible to pass a table pointer or NULL (indicating the use of the
1034: internal tables) to \fBpcre_exec()\fP. Although not intended for this purpose,
1035: this facility could be used to match a pattern in a different locale from the
1036: one in which it was compiled. Passing table pointers at run time is discussed
1037: below in the section on matching a pattern.
1038: .
1039: .
1040: .\" HTML <a name="infoaboutpattern"></a>
1041: .SH "INFORMATION ABOUT A PATTERN"
1042: .rs
1043: .sp
1044: .B int pcre_fullinfo(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
1045: .ti +5n
1046: .B int \fIwhat\fP, void *\fIwhere\fP);
1047: .PP
1048: The \fBpcre_fullinfo()\fP function returns information about a compiled
1049: pattern. It replaces the obsolete \fBpcre_info()\fP function, which is
1050: nevertheless retained for backwards compability (and is documented below).
1051: .P
1052: The first argument for \fBpcre_fullinfo()\fP is a pointer to the compiled
1053: pattern. The second argument is the result of \fBpcre_study()\fP, or NULL if
1054: the pattern was not studied. The third argument specifies which piece of
1055: information is required, and the fourth argument is a pointer to a variable
1056: to receive the data. The yield of the function is zero for success, or one of
1057: the following negative numbers:
1058: .sp
1059: PCRE_ERROR_NULL the argument \fIcode\fP was NULL
1060: the argument \fIwhere\fP was NULL
1061: PCRE_ERROR_BADMAGIC the "magic number" was not found
1062: PCRE_ERROR_BADOPTION the value of \fIwhat\fP was invalid
1063: .sp
1064: The "magic number" is placed at the start of each compiled pattern as an simple
1065: check against passing an arbitrary memory pointer. Here is a typical call of
1066: \fBpcre_fullinfo()\fP, to obtain the length of the compiled pattern:
1067: .sp
1068: int rc;
1069: size_t length;
1070: rc = pcre_fullinfo(
1071: re, /* result of pcre_compile() */
1072: sd, /* result of pcre_study(), or NULL */
1073: PCRE_INFO_SIZE, /* what is required */
1074: &length); /* where to put the data */
1075: .sp
1076: The possible values for the third argument are defined in \fBpcre.h\fP, and are
1077: as follows:
1078: .sp
1079: PCRE_INFO_BACKREFMAX
1080: .sp
1081: Return the number of the highest back reference in the pattern. The fourth
1082: argument should point to an \fBint\fP variable. Zero is returned if there are
1083: no back references.
1084: .sp
1085: PCRE_INFO_CAPTURECOUNT
1086: .sp
1087: Return the number of capturing subpatterns in the pattern. The fourth argument
1088: should point to an \fBint\fP variable.
1089: .sp
1090: PCRE_INFO_DEFAULT_TABLES
1091: .sp
1092: Return a pointer to the internal default character tables within PCRE. The
1093: fourth argument should point to an \fBunsigned char *\fP variable. This
1094: information call is provided for internal use by the \fBpcre_study()\fP
1095: function. External callers can cause PCRE to use its internal tables by passing
1096: a NULL table pointer.
1097: .sp
1098: PCRE_INFO_FIRSTBYTE
1099: .sp
1100: Return information about the first byte of any matched string, for a
1101: non-anchored pattern. The fourth argument should point to an \fBint\fP
1102: variable. (This option used to be called PCRE_INFO_FIRSTCHAR; the old name is
1103: still recognized for backwards compatibility.)
1104: .P
1105: If there is a fixed first byte, for example, from a pattern such as
1106: (cat|cow|coyote), its value is returned. Otherwise, if either
1107: .sp
1108: (a) the pattern was compiled with the PCRE_MULTILINE option, and every branch
1109: starts with "^", or
1110: .sp
1111: (b) every branch of the pattern starts with ".*" and PCRE_DOTALL is not set
1112: (if it were set, the pattern would be anchored),
1113: .sp
1114: -1 is returned, indicating that the pattern matches only at the start of a
1115: subject string or after any newline within the string. Otherwise -2 is
1116: returned. For anchored patterns, -2 is returned.
1117: .sp
1118: PCRE_INFO_FIRSTTABLE
1119: .sp
1120: If the pattern was studied, and this resulted in the construction of a 256-bit
1121: table indicating a fixed set of bytes for the first byte in any matching
1122: string, a pointer to the table is returned. Otherwise NULL is returned. The
1123: fourth argument should point to an \fBunsigned char *\fP variable.
1124: .sp
1125: PCRE_INFO_HASCRORLF
1126: .sp
1127: Return 1 if the pattern contains any explicit matches for CR or LF characters,
1128: otherwise 0. The fourth argument should point to an \fBint\fP variable. An
1129: explicit match is either a literal CR or LF character, or \er or \en.
1130: .sp
1131: PCRE_INFO_JCHANGED
1132: .sp
1133: Return 1 if the (?J) or (?-J) option setting is used in the pattern, otherwise
1134: 0. The fourth argument should point to an \fBint\fP variable. (?J) and
1135: (?-J) set and unset the local PCRE_DUPNAMES option, respectively.
1136: .sp
1137: PCRE_INFO_JIT
1138: .sp
1139: Return 1 if the pattern was studied with the PCRE_STUDY_JIT_COMPILE option, and
1140: just-in-time compiling was successful. The fourth argument should point to an
1141: \fBint\fP variable. A return value of 0 means that JIT support is not available
1142: in this version of PCRE, or that the pattern was not studied with the
1143: PCRE_STUDY_JIT_COMPILE option, or that the JIT compiler could not handle this
1144: particular pattern. See the
1145: .\" HREF
1146: \fBpcrejit\fP
1147: .\"
1148: documentation for details of what can and cannot be handled.
1149: .sp
1150: PCRE_INFO_JITSIZE
1151: .sp
1152: If the pattern was successfully studied with the PCRE_STUDY_JIT_COMPILE option,
1153: return the size of the JIT compiled code, otherwise return zero. The fourth
1154: argument should point to a \fBsize_t\fP variable.
1155: .sp
1156: PCRE_INFO_LASTLITERAL
1157: .sp
1158: Return the value of the rightmost literal byte that must exist in any matched
1159: string, other than at its start, if such a byte has been recorded. The fourth
1160: argument should point to an \fBint\fP variable. If there is no such byte, -1 is
1161: returned. For anchored patterns, a last literal byte is recorded only if it
1162: follows something of variable length. For example, for the pattern
1163: /^a\ed+z\ed+/ the returned value is "z", but for /^a\edz\ed/ the returned value
1164: is -1.
1165: .sp
1166: PCRE_INFO_MINLENGTH
1167: .sp
1168: If the pattern was studied and a minimum length for matching subject strings
1169: was computed, its value is returned. Otherwise the returned value is -1. The
1170: value is a number of characters, not bytes (this may be relevant in UTF-8
1171: mode). The fourth argument should point to an \fBint\fP variable. A
1172: non-negative value is a lower bound to the length of any matching string. There
1173: may not be any strings of that length that do actually match, but every string
1174: that does match is at least that long.
1175: .sp
1176: PCRE_INFO_NAMECOUNT
1177: PCRE_INFO_NAMEENTRYSIZE
1178: PCRE_INFO_NAMETABLE
1179: .sp
1180: PCRE supports the use of named as well as numbered capturing parentheses. The
1181: names are just an additional way of identifying the parentheses, which still
1182: acquire numbers. Several convenience functions such as
1183: \fBpcre_get_named_substring()\fP are provided for extracting captured
1184: substrings by name. It is also possible to extract the data directly, by first
1185: converting the name to a number in order to access the correct pointers in the
1186: output vector (described with \fBpcre_exec()\fP below). To do the conversion,
1187: you need to use the name-to-number map, which is described by these three
1188: values.
1189: .P
1190: The map consists of a number of fixed-size entries. PCRE_INFO_NAMECOUNT gives
1191: the number of entries, and PCRE_INFO_NAMEENTRYSIZE gives the size of each
1192: entry; both of these return an \fBint\fP value. The entry size depends on the
1193: length of the longest name. PCRE_INFO_NAMETABLE returns a pointer to the first
1194: entry of the table (a pointer to \fBchar\fP). The first two bytes of each entry
1195: are the number of the capturing parenthesis, most significant byte first. The
1196: rest of the entry is the corresponding name, zero terminated.
1197: .P
1198: The names are in alphabetical order. Duplicate names may appear if (?| is used
1199: to create multiple groups with the same number, as described in the
1200: .\" HTML <a href="pcrepattern.html#dupsubpatternnumber">
1201: .\" </a>
1202: section on duplicate subpattern numbers
1203: .\"
1204: in the
1205: .\" HREF
1206: \fBpcrepattern\fP
1207: .\"
1208: page. Duplicate names for subpatterns with different numbers are permitted only
1209: if PCRE_DUPNAMES is set. In all cases of duplicate names, they appear in the
1210: table in the order in which they were found in the pattern. In the absence of
1211: (?| this is the order of increasing number; when (?| is used this is not
1212: necessarily the case because later subpatterns may have lower numbers.
1213: .P
1214: As a simple example of the name/number table, consider the following pattern
1215: (assume PCRE_EXTENDED is set, so white space - including newlines - is
1216: ignored):
1217: .sp
1218: .\" JOIN
1219: (?<date> (?<year>(\ed\ed)?\ed\ed) -
1220: (?<month>\ed\ed) - (?<day>\ed\ed) )
1221: .sp
1222: There are four named subpatterns, so the table has four entries, and each entry
1223: in the table is eight bytes long. The table is as follows, with non-printing
1224: bytes shows in hexadecimal, and undefined bytes shown as ??:
1225: .sp
1226: 00 01 d a t e 00 ??
1227: 00 05 d a y 00 ?? ??
1228: 00 04 m o n t h 00
1229: 00 02 y e a r 00 ??
1230: .sp
1231: When writing code to extract data from named subpatterns using the
1232: name-to-number map, remember that the length of the entries is likely to be
1233: different for each compiled pattern.
1234: .sp
1235: PCRE_INFO_OKPARTIAL
1236: .sp
1237: Return 1 if the pattern can be used for partial matching with
1238: \fBpcre_exec()\fP, otherwise 0. The fourth argument should point to an
1239: \fBint\fP variable. From release 8.00, this always returns 1, because the
1240: restrictions that previously applied to partial matching have been lifted. The
1241: .\" HREF
1242: \fBpcrepartial\fP
1243: .\"
1244: documentation gives details of partial matching.
1245: .sp
1246: PCRE_INFO_OPTIONS
1247: .sp
1248: Return a copy of the options with which the pattern was compiled. The fourth
1249: argument should point to an \fBunsigned long int\fP variable. These option bits
1250: are those specified in the call to \fBpcre_compile()\fP, modified by any
1251: top-level option settings at the start of the pattern itself. In other words,
1252: they are the options that will be in force when matching starts. For example,
1253: if the pattern /(?im)abc(?-i)d/ is compiled with the PCRE_EXTENDED option, the
1254: result is PCRE_CASELESS, PCRE_MULTILINE, and PCRE_EXTENDED.
1255: .P
1256: A pattern is automatically anchored by PCRE if all of its top-level
1257: alternatives begin with one of the following:
1258: .sp
1259: ^ unless PCRE_MULTILINE is set
1260: \eA always
1261: \eG always
1262: .\" JOIN
1263: .* if PCRE_DOTALL is set and there are no back
1264: references to the subpattern in which .* appears
1265: .sp
1266: For such patterns, the PCRE_ANCHORED bit is set in the options returned by
1267: \fBpcre_fullinfo()\fP.
1268: .sp
1269: PCRE_INFO_SIZE
1270: .sp
1271: Return the size of the compiled pattern. The fourth argument should point to a
1272: \fBsize_t\fP variable. This value does not include the size of the \fBpcre\fP
1273: structure that is returned by \fBpcre_compile()\fP. The value that is passed as
1274: the argument to \fBpcre_malloc()\fP when \fBpcre_compile()\fP is getting memory
1275: in which to place the compiled data is the value returned by this option plus
1276: the size of the \fBpcre\fP structure. Studying a compiled pattern, with or
1277: without JIT, does not alter the value returned by this option.
1278: .sp
1279: PCRE_INFO_STUDYSIZE
1280: .sp
1281: Return the size of the data block pointed to by the \fIstudy_data\fP field in a
1282: \fBpcre_extra\fP block. If \fBpcre_extra\fP is NULL, or there is no study data,
1283: zero is returned. The fourth argument should point to a \fBsize_t\fP variable.
1284: The \fIstudy_data\fP field is set by \fBpcre_study()\fP to record information
1285: that will speed up matching (see the section entitled
1286: .\" HTML <a href="#studyingapattern">
1287: .\" </a>
1288: "Studying a pattern"
1289: .\"
1290: above). The format of the \fIstudy_data\fP block is private, but its length
1291: is made available via this option so that it can be saved and restored (see the
1292: .\" HREF
1293: \fBpcreprecompile\fP
1294: .\"
1295: documentation for details).
1296: .
1297: .
1298: .SH "OBSOLETE INFO FUNCTION"
1299: .rs
1300: .sp
1301: .B int pcre_info(const pcre *\fIcode\fP, int *\fIoptptr\fP, int
1302: .B *\fIfirstcharptr\fP);
1303: .PP
1304: The \fBpcre_info()\fP function is now obsolete because its interface is too
1305: restrictive to return all the available data about a compiled pattern. New
1306: programs should use \fBpcre_fullinfo()\fP instead. The yield of
1307: \fBpcre_info()\fP is the number of capturing subpatterns, or one of the
1308: following negative numbers:
1309: .sp
1310: PCRE_ERROR_NULL the argument \fIcode\fP was NULL
1311: PCRE_ERROR_BADMAGIC the "magic number" was not found
1312: .sp
1313: If the \fIoptptr\fP argument is not NULL, a copy of the options with which the
1314: pattern was compiled is placed in the integer it points to (see
1315: PCRE_INFO_OPTIONS above).
1316: .P
1317: If the pattern is not anchored and the \fIfirstcharptr\fP argument is not NULL,
1318: it is used to pass back information about the first character of any matched
1319: string (see PCRE_INFO_FIRSTBYTE above).
1320: .
1321: .
1322: .SH "REFERENCE COUNTS"
1323: .rs
1324: .sp
1325: .B int pcre_refcount(pcre *\fIcode\fP, int \fIadjust\fP);
1326: .PP
1327: The \fBpcre_refcount()\fP function is used to maintain a reference count in the
1328: data block that contains a compiled pattern. It is provided for the benefit of
1329: applications that operate in an object-oriented manner, where different parts
1330: of the application may be using the same compiled pattern, but you want to free
1331: the block when they are all done.
1332: .P
1333: When a pattern is compiled, the reference count field is initialized to zero.
1334: It is changed only by calling this function, whose action is to add the
1335: \fIadjust\fP value (which may be positive or negative) to it. The yield of the
1336: function is the new value. However, the value of the count is constrained to
1337: lie between 0 and 65535, inclusive. If the new value is outside these limits,
1338: it is forced to the appropriate limit value.
1339: .P
1340: Except when it is zero, the reference count is not correctly preserved if a
1341: pattern is compiled on one host and then transferred to a host whose byte-order
1342: is different. (This seems a highly unlikely scenario.)
1343: .
1344: .
1345: .SH "MATCHING A PATTERN: THE TRADITIONAL FUNCTION"
1346: .rs
1347: .sp
1348: .B int pcre_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
1349: .ti +5n
1350: .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
1351: .ti +5n
1352: .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP);
1353: .P
1354: The function \fBpcre_exec()\fP is called to match a subject string against a
1355: compiled pattern, which is passed in the \fIcode\fP argument. If the
1356: pattern was studied, the result of the study should be passed in the
1357: \fIextra\fP argument. You can call \fBpcre_exec()\fP with the same \fIcode\fP
1358: and \fIextra\fP arguments as many times as you like, in order to match
1359: different subject strings with the same pattern.
1360: .P
1361: This function is the main matching facility of the library, and it operates in
1362: a Perl-like manner. For specialist use there is also an alternative matching
1363: function, which is described
1364: .\" HTML <a href="#dfamatch">
1365: .\" </a>
1366: below
1367: .\"
1368: in the section about the \fBpcre_dfa_exec()\fP function.
1369: .P
1370: In most applications, the pattern will have been compiled (and optionally
1371: studied) in the same process that calls \fBpcre_exec()\fP. However, it is
1372: possible to save compiled patterns and study data, and then use them later
1373: in different processes, possibly even on different hosts. For a discussion
1374: about this, see the
1375: .\" HREF
1376: \fBpcreprecompile\fP
1377: .\"
1378: documentation.
1379: .P
1380: Here is an example of a simple call to \fBpcre_exec()\fP:
1381: .sp
1382: int rc;
1383: int ovector[30];
1384: rc = pcre_exec(
1385: re, /* result of pcre_compile() */
1386: NULL, /* we didn't study the pattern */
1387: "some string", /* the subject string */
1388: 11, /* the length of the subject string */
1389: 0, /* start at offset 0 in the subject */
1390: 0, /* default options */
1391: ovector, /* vector of integers for substring information */
1392: 30); /* number of elements (NOT size in bytes) */
1393: .
1394: .
1395: .\" HTML <a name="extradata"></a>
1396: .SS "Extra data for \fBpcre_exec()\fR"
1397: .rs
1398: .sp
1399: If the \fIextra\fP argument is not NULL, it must point to a \fBpcre_extra\fP
1400: data block. The \fBpcre_study()\fP function returns such a block (when it
1401: doesn't return NULL), but you can also create one for yourself, and pass
1402: additional information in it. The \fBpcre_extra\fP block contains the following
1403: fields (not necessarily in this order):
1404: .sp
1405: unsigned long int \fIflags\fP;
1406: void *\fIstudy_data\fP;
1407: void *\fIexecutable_jit\fP;
1408: unsigned long int \fImatch_limit\fP;
1409: unsigned long int \fImatch_limit_recursion\fP;
1410: void *\fIcallout_data\fP;
1411: const unsigned char *\fItables\fP;
1412: unsigned char **\fImark\fP;
1413: .sp
1414: The \fIflags\fP field is a bitmap that specifies which of the other fields
1415: are set. The flag bits are:
1416: .sp
1417: PCRE_EXTRA_STUDY_DATA
1418: PCRE_EXTRA_EXECUTABLE_JIT
1419: PCRE_EXTRA_MATCH_LIMIT
1420: PCRE_EXTRA_MATCH_LIMIT_RECURSION
1421: PCRE_EXTRA_CALLOUT_DATA
1422: PCRE_EXTRA_TABLES
1423: PCRE_EXTRA_MARK
1424: .sp
1425: Other flag bits should be set to zero. The \fIstudy_data\fP field and sometimes
1426: the \fIexecutable_jit\fP field are set in the \fBpcre_extra\fP block that is
1427: returned by \fBpcre_study()\fP, together with the appropriate flag bits. You
1428: should not set these yourself, but you may add to the block by setting the
1429: other fields and their corresponding flag bits.
1430: .P
1431: The \fImatch_limit\fP field provides a means of preventing PCRE from using up a
1432: vast amount of resources when running patterns that are not going to match,
1433: but which have a very large number of possibilities in their search trees. The
1434: classic example is a pattern that uses nested unlimited repeats.
1435: .P
1436: Internally, \fBpcre_exec()\fP uses a function called \fBmatch()\fP, which it
1437: calls repeatedly (sometimes recursively). The limit set by \fImatch_limit\fP is
1438: imposed on the number of times this function is called during a match, which
1439: has the effect of limiting the amount of backtracking that can take place. For
1440: patterns that are not anchored, the count restarts from zero for each position
1441: in the subject string.
1442: .P
1443: When \fBpcre_exec()\fP is called with a pattern that was successfully studied
1444: with the PCRE_STUDY_JIT_COMPILE option, the way that the matching is executed
1445: is entirely different. However, there is still the possibility of runaway
1446: matching that goes on for a very long time, and so the \fImatch_limit\fP value
1447: is also used in this case (but in a different way) to limit how long the
1448: matching can continue.
1449: .P
1450: The default value for the limit can be set when PCRE is built; the default
1451: default is 10 million, which handles all but the most extreme cases. You can
1452: override the default by suppling \fBpcre_exec()\fP with a \fBpcre_extra\fP
1453: block in which \fImatch_limit\fP is set, and PCRE_EXTRA_MATCH_LIMIT is set in
1454: the \fIflags\fP field. If the limit is exceeded, \fBpcre_exec()\fP returns
1455: PCRE_ERROR_MATCHLIMIT.
1456: .P
1457: The \fImatch_limit_recursion\fP field is similar to \fImatch_limit\fP, but
1458: instead of limiting the total number of times that \fBmatch()\fP is called, it
1459: limits the depth of recursion. The recursion depth is a smaller number than the
1460: total number of calls, because not all calls to \fBmatch()\fP are recursive.
1461: This limit is of use only if it is set smaller than \fImatch_limit\fP.
1462: .P
1463: Limiting the recursion depth limits the amount of machine stack that can be
1464: used, or, when PCRE has been compiled to use memory on the heap instead of the
1465: stack, the amount of heap memory that can be used. This limit is not relevant,
1466: and is ignored, if the pattern was successfully studied with
1467: PCRE_STUDY_JIT_COMPILE.
1468: .P
1469: The default value for \fImatch_limit_recursion\fP can be set when PCRE is
1470: built; the default default is the same value as the default for
1471: \fImatch_limit\fP. You can override the default by suppling \fBpcre_exec()\fP
1472: with a \fBpcre_extra\fP block in which \fImatch_limit_recursion\fP is set, and
1473: PCRE_EXTRA_MATCH_LIMIT_RECURSION is set in the \fIflags\fP field. If the limit
1474: is exceeded, \fBpcre_exec()\fP returns PCRE_ERROR_RECURSIONLIMIT.
1475: .P
1476: The \fIcallout_data\fP field is used in conjunction with the "callout" feature,
1477: and is described in the
1478: .\" HREF
1479: \fBpcrecallout\fP
1480: .\"
1481: documentation.
1482: .P
1483: The \fItables\fP field is used to pass a character tables pointer to
1484: \fBpcre_exec()\fP; this overrides the value that is stored with the compiled
1485: pattern. A non-NULL value is stored with the compiled pattern only if custom
1486: tables were supplied to \fBpcre_compile()\fP via its \fItableptr\fP argument.
1487: If NULL is passed to \fBpcre_exec()\fP using this mechanism, it forces PCRE's
1488: internal tables to be used. This facility is helpful when re-using patterns
1489: that have been saved after compiling with an external set of tables, because
1490: the external tables might be at a different address when \fBpcre_exec()\fP is
1491: called. See the
1492: .\" HREF
1493: \fBpcreprecompile\fP
1494: .\"
1495: documentation for a discussion of saving compiled patterns for later use.
1496: .P
1497: If PCRE_EXTRA_MARK is set in the \fIflags\fP field, the \fImark\fP field must
1498: be set to point to a \fBchar *\fP variable. If the pattern contains any
1499: backtracking control verbs such as (*MARK:NAME), and the execution ends up with
1500: a name to pass back, a pointer to the name string (zero terminated) is placed
1501: in the variable pointed to by the \fImark\fP field. The names are within the
1502: compiled pattern; if you wish to retain such a name you must copy it before
1503: freeing the memory of a compiled pattern. If there is no name to pass back, the
1504: variable pointed to by the \fImark\fP field set to NULL. For details of the
1505: backtracking control verbs, see the section entitled
1506: .\" HTML <a href="pcrepattern#backtrackcontrol">
1507: .\" </a>
1508: "Backtracking control"
1509: .\"
1510: in the
1511: .\" HREF
1512: \fBpcrepattern\fP
1513: .\"
1514: documentation.
1515: .
1516: .
1517: .\" HTML <a name="execoptions"></a>
1518: .SS "Option bits for \fBpcre_exec()\fP"
1519: .rs
1520: .sp
1521: The unused bits of the \fIoptions\fP argument for \fBpcre_exec()\fP must be
1522: zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_\fIxxx\fP,
1523: PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NOTEMPTY_ATSTART,
1524: PCRE_NO_START_OPTIMIZE, PCRE_NO_UTF8_CHECK, PCRE_PARTIAL_SOFT, and
1525: PCRE_PARTIAL_HARD.
1526: .P
1527: If the pattern was successfully studied with the PCRE_STUDY_JIT_COMPILE option,
1528: the only supported options for JIT execution are PCRE_NO_UTF8_CHECK,
1529: PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, and PCRE_NOTEMPTY_ATSTART. Note in
1530: particular that partial matching is not supported. If an unsupported option is
1531: used, JIT execution is disabled and the normal interpretive code in
1532: \fBpcre_exec()\fP is run.
1533: .sp
1534: PCRE_ANCHORED
1535: .sp
1536: The PCRE_ANCHORED option limits \fBpcre_exec()\fP to matching at the first
1537: matching position. If a pattern was compiled with PCRE_ANCHORED, or turned out
1538: to be anchored by virtue of its contents, it cannot be made unachored at
1539: matching time.
1540: .sp
1541: PCRE_BSR_ANYCRLF
1542: PCRE_BSR_UNICODE
1543: .sp
1544: These options (which are mutually exclusive) control what the \eR escape
1545: sequence matches. The choice is either to match only CR, LF, or CRLF, or to
1546: match any Unicode newline sequence. These options override the choice that was
1547: made or defaulted when the pattern was compiled.
1548: .sp
1549: PCRE_NEWLINE_CR
1550: PCRE_NEWLINE_LF
1551: PCRE_NEWLINE_CRLF
1552: PCRE_NEWLINE_ANYCRLF
1553: PCRE_NEWLINE_ANY
1554: .sp
1555: These options override the newline definition that was chosen or defaulted when
1556: the pattern was compiled. For details, see the description of
1557: \fBpcre_compile()\fP above. During matching, the newline choice affects the
1558: behaviour of the dot, circumflex, and dollar metacharacters. It may also alter
1559: the way the match position is advanced after a match failure for an unanchored
1560: pattern.
1561: .P
1562: When PCRE_NEWLINE_CRLF, PCRE_NEWLINE_ANYCRLF, or PCRE_NEWLINE_ANY is set, and a
1563: match attempt for an unanchored pattern fails when the current position is at a
1564: CRLF sequence, and the pattern contains no explicit matches for CR or LF
1565: characters, the match position is advanced by two characters instead of one, in
1566: other words, to after the CRLF.
1567: .P
1568: The above rule is a compromise that makes the most common cases work as
1569: expected. For example, if the pattern is .+A (and the PCRE_DOTALL option is not
1570: set), it does not match the string "\er\enA" because, after failing at the
1571: start, it skips both the CR and the LF before retrying. However, the pattern
1572: [\er\en]A does match that string, because it contains an explicit CR or LF
1573: reference, and so advances only by one character after the first failure.
1574: .P
1575: An explicit match for CR of LF is either a literal appearance of one of those
1576: characters, or one of the \er or \en escape sequences. Implicit matches such as
1577: [^X] do not count, nor does \es (which includes CR and LF in the characters
1578: that it matches).
1579: .P
1580: Notwithstanding the above, anomalous effects may still occur when CRLF is a
1581: valid newline sequence and explicit \er or \en escapes appear in the pattern.
1582: .sp
1583: PCRE_NOTBOL
1584: .sp
1585: This option specifies that first character of the subject string is not the
1586: beginning of a line, so the circumflex metacharacter should not match before
1587: it. Setting this without PCRE_MULTILINE (at compile time) causes circumflex
1588: never to match. This option affects only the behaviour of the circumflex
1589: metacharacter. It does not affect \eA.
1590: .sp
1591: PCRE_NOTEOL
1592: .sp
1593: This option specifies that the end of the subject string is not the end of a
1594: line, so the dollar metacharacter should not match it nor (except in multiline
1595: mode) a newline immediately before it. Setting this without PCRE_MULTILINE (at
1596: compile time) causes dollar never to match. This option affects only the
1597: behaviour of the dollar metacharacter. It does not affect \eZ or \ez.
1598: .sp
1599: PCRE_NOTEMPTY
1600: .sp
1601: An empty string is not considered to be a valid match if this option is set. If
1602: there are alternatives in the pattern, they are tried. If all the alternatives
1603: match the empty string, the entire match fails. For example, if the pattern
1604: .sp
1605: a?b?
1606: .sp
1607: is applied to a string not beginning with "a" or "b", it matches an empty
1608: string at the start of the subject. With PCRE_NOTEMPTY set, this match is not
1609: valid, so PCRE searches further into the string for occurrences of "a" or "b".
1610: .sp
1611: PCRE_NOTEMPTY_ATSTART
1612: .sp
1613: This is like PCRE_NOTEMPTY, except that an empty string match that is not at
1614: the start of the subject is permitted. If the pattern is anchored, such a match
1615: can occur only if the pattern contains \eK.
1616: .P
1617: Perl has no direct equivalent of PCRE_NOTEMPTY or PCRE_NOTEMPTY_ATSTART, but it
1618: does make a special case of a pattern match of the empty string within its
1619: \fBsplit()\fP function, and when using the /g modifier. It is possible to
1620: emulate Perl's behaviour after matching a null string by first trying the match
1621: again at the same offset with PCRE_NOTEMPTY_ATSTART and PCRE_ANCHORED, and then
1622: if that fails, by advancing the starting offset (see below) and trying an
1623: ordinary match again. There is some code that demonstrates how to do this in
1624: the
1625: .\" HREF
1626: \fBpcredemo\fP
1627: .\"
1628: sample program. In the most general case, you have to check to see if the
1629: newline convention recognizes CRLF as a newline, and if so, and the current
1630: character is CR followed by LF, advance the starting offset by two characters
1631: instead of one.
1632: .sp
1633: PCRE_NO_START_OPTIMIZE
1634: .sp
1635: There are a number of optimizations that \fBpcre_exec()\fP uses at the start of
1636: a match, in order to speed up the process. For example, if it is known that an
1637: unanchored match must start with a specific character, it searches the subject
1638: for that character, and fails immediately if it cannot find it, without
1639: actually running the main matching function. This means that a special item
1640: such as (*COMMIT) at the start of a pattern is not considered until after a
1641: suitable starting point for the match has been found. When callouts or (*MARK)
1642: items are in use, these "start-up" optimizations can cause them to be skipped
1643: if the pattern is never actually used. The start-up optimizations are in effect
1644: a pre-scan of the subject that takes place before the pattern is run.
1645: .P
1646: The PCRE_NO_START_OPTIMIZE option disables the start-up optimizations, possibly
1647: causing performance to suffer, but ensuring that in cases where the result is
1648: "no match", the callouts do occur, and that items such as (*COMMIT) and (*MARK)
1649: are considered at every possible starting position in the subject string. If
1650: PCRE_NO_START_OPTIMIZE is set at compile time, it cannot be unset at matching
1651: time.
1652: .P
1653: Setting PCRE_NO_START_OPTIMIZE can change the outcome of a matching operation.
1654: Consider the pattern
1655: .sp
1656: (*COMMIT)ABC
1657: .sp
1658: When this is compiled, PCRE records the fact that a match must start with the
1659: character "A". Suppose the subject string is "DEFABC". The start-up
1660: optimization scans along the subject, finds "A" and runs the first match
1661: attempt from there. The (*COMMIT) item means that the pattern must match the
1662: current starting position, which in this case, it does. However, if the same
1663: match is run with PCRE_NO_START_OPTIMIZE set, the initial scan along the
1664: subject string does not happen. The first match attempt is run starting from
1665: "D" and when this fails, (*COMMIT) prevents any further matches being tried, so
1666: the overall result is "no match". If the pattern is studied, more start-up
1667: optimizations may be used. For example, a minimum length for the subject may be
1668: recorded. Consider the pattern
1669: .sp
1670: (*MARK:A)(X|Y)
1671: .sp
1672: The minimum length for a match is one character. If the subject is "ABC", there
1673: will be attempts to match "ABC", "BC", "C", and then finally an empty string.
1674: If the pattern is studied, the final attempt does not take place, because PCRE
1675: knows that the subject is too short, and so the (*MARK) is never encountered.
1676: In this case, studying the pattern does not affect the overall match result,
1677: which is still "no match", but it does affect the auxiliary information that is
1678: returned.
1679: .sp
1680: PCRE_NO_UTF8_CHECK
1681: .sp
1682: When PCRE_UTF8 is set at compile time, the validity of the subject as a UTF-8
1683: string is automatically checked when \fBpcre_exec()\fP is subsequently called.
1684: The value of \fIstartoffset\fP is also checked to ensure that it points to the
1685: start of a UTF-8 character. There is a discussion about the validity of UTF-8
1686: strings in the
1687: .\" HTML <a href="pcre.html#utf8strings">
1688: .\" </a>
1689: section on UTF-8 support
1690: .\"
1691: in the main
1692: .\" HREF
1693: \fBpcre\fP
1694: .\"
1695: page. If an invalid UTF-8 sequence of bytes is found, \fBpcre_exec()\fP returns
1696: the error PCRE_ERROR_BADUTF8 or, if PCRE_PARTIAL_HARD is set and the problem is
1697: a truncated UTF-8 character at the end of the subject, PCRE_ERROR_SHORTUTF8. In
1698: both cases, information about the precise nature of the error may also be
1699: returned (see the descriptions of these errors in the section entitled \fIError
1700: return values from\fP \fBpcre_exec()\fP
1701: .\" HTML <a href="#errorlist">
1702: .\" </a>
1703: below).
1704: .\"
1705: If \fIstartoffset\fP contains a value that does not point to the start of a
1706: UTF-8 character (or to the end of the subject), PCRE_ERROR_BADUTF8_OFFSET is
1707: returned.
1708: .P
1709: If you already know that your subject is valid, and you want to skip these
1710: checks for performance reasons, you can set the PCRE_NO_UTF8_CHECK option when
1711: calling \fBpcre_exec()\fP. You might want to do this for the second and
1712: subsequent calls to \fBpcre_exec()\fP if you are making repeated calls to find
1713: all the matches in a single subject string. However, you should be sure that
1714: the value of \fIstartoffset\fP points to the start of a UTF-8 character (or the
1715: end of the subject). When PCRE_NO_UTF8_CHECK is set, the effect of passing an
1716: invalid UTF-8 string as a subject or an invalid value of \fIstartoffset\fP is
1717: undefined. Your program may crash.
1718: .sp
1719: PCRE_PARTIAL_HARD
1720: PCRE_PARTIAL_SOFT
1721: .sp
1722: These options turn on the partial matching feature. For backwards
1723: compatibility, PCRE_PARTIAL is a synonym for PCRE_PARTIAL_SOFT. A partial match
1724: occurs if the end of the subject string is reached successfully, but there are
1725: not enough subject characters to complete the match. If this happens when
1726: PCRE_PARTIAL_SOFT (but not PCRE_PARTIAL_HARD) is set, matching continues by
1727: testing any remaining alternatives. Only if no complete match can be found is
1728: PCRE_ERROR_PARTIAL returned instead of PCRE_ERROR_NOMATCH. In other words,
1729: PCRE_PARTIAL_SOFT says that the caller is prepared to handle a partial match,
1730: but only if no complete match can be found.
1731: .P
1732: If PCRE_PARTIAL_HARD is set, it overrides PCRE_PARTIAL_SOFT. In this case, if a
1733: partial match is found, \fBpcre_exec()\fP immediately returns
1734: PCRE_ERROR_PARTIAL, without considering any other alternatives. In other words,
1735: when PCRE_PARTIAL_HARD is set, a partial match is considered to be more
1736: important that an alternative complete match.
1737: .P
1738: In both cases, the portion of the string that was inspected when the partial
1739: match was found is set as the first matching string. There is a more detailed
1740: discussion of partial and multi-segment matching, with examples, in the
1741: .\" HREF
1742: \fBpcrepartial\fP
1743: .\"
1744: documentation.
1745: .
1746: .
1747: .SS "The string to be matched by \fBpcre_exec()\fP"
1748: .rs
1749: .sp
1750: The subject string is passed to \fBpcre_exec()\fP as a pointer in
1751: \fIsubject\fP, a length (in bytes) in \fIlength\fP, and a starting byte offset
1752: in \fIstartoffset\fP. If this is negative or greater than the length of the
1753: subject, \fBpcre_exec()\fP returns PCRE_ERROR_BADOFFSET. When the starting
1754: offset is zero, the search for a match starts at the beginning of the subject,
1755: and this is by far the most common case. In UTF-8 mode, the byte offset must
1756: point to the start of a UTF-8 character (or the end of the subject). Unlike the
1757: pattern string, the subject may contain binary zero bytes.
1758: .P
1759: A non-zero starting offset is useful when searching for another match in the
1760: same subject by calling \fBpcre_exec()\fP again after a previous success.
1761: Setting \fIstartoffset\fP differs from just passing over a shortened string and
1762: setting PCRE_NOTBOL in the case of a pattern that begins with any kind of
1763: lookbehind. For example, consider the pattern
1764: .sp
1765: \eBiss\eB
1766: .sp
1767: which finds occurrences of "iss" in the middle of words. (\eB matches only if
1768: the current position in the subject is not a word boundary.) When applied to
1769: the string "Mississipi" the first call to \fBpcre_exec()\fP finds the first
1770: occurrence. If \fBpcre_exec()\fP is called again with just the remainder of the
1771: subject, namely "issipi", it does not match, because \eB is always false at the
1772: start of the subject, which is deemed to be a word boundary. However, if
1773: \fBpcre_exec()\fP is passed the entire string again, but with \fIstartoffset\fP
1774: set to 4, it finds the second occurrence of "iss" because it is able to look
1775: behind the starting point to discover that it is preceded by a letter.
1776: .P
1777: Finding all the matches in a subject is tricky when the pattern can match an
1778: empty string. It is possible to emulate Perl's /g behaviour by first trying the
1779: match again at the same offset, with the PCRE_NOTEMPTY_ATSTART and
1780: PCRE_ANCHORED options, and then if that fails, advancing the starting offset
1781: and trying an ordinary match again. There is some code that demonstrates how to
1782: do this in the
1783: .\" HREF
1784: \fBpcredemo\fP
1785: .\"
1786: sample program. In the most general case, you have to check to see if the
1787: newline convention recognizes CRLF as a newline, and if so, and the current
1788: character is CR followed by LF, advance the starting offset by two characters
1789: instead of one.
1790: .P
1791: If a non-zero starting offset is passed when the pattern is anchored, one
1792: attempt to match at the given offset is made. This can only succeed if the
1793: pattern does not require the match to be at the start of the subject.
1794: .
1795: .
1796: .SS "How \fBpcre_exec()\fP returns captured substrings"
1797: .rs
1798: .sp
1799: In general, a pattern matches a certain portion of the subject, and in
1800: addition, further substrings from the subject may be picked out by parts of the
1801: pattern. Following the usage in Jeffrey Friedl's book, this is called
1802: "capturing" in what follows, and the phrase "capturing subpattern" is used for
1803: a fragment of a pattern that picks out a substring. PCRE supports several other
1804: kinds of parenthesized subpattern that do not cause substrings to be captured.
1805: .P
1806: Captured substrings are returned to the caller via a vector of integers whose
1807: address is passed in \fIovector\fP. The number of elements in the vector is
1808: passed in \fIovecsize\fP, which must be a non-negative number. \fBNote\fP: this
1809: argument is NOT the size of \fIovector\fP in bytes.
1810: .P
1811: The first two-thirds of the vector is used to pass back captured substrings,
1812: each substring using a pair of integers. The remaining third of the vector is
1813: used as workspace by \fBpcre_exec()\fP while matching capturing subpatterns,
1814: and is not available for passing back information. The number passed in
1815: \fIovecsize\fP should always be a multiple of three. If it is not, it is
1816: rounded down.
1817: .P
1818: When a match is successful, information about captured substrings is returned
1819: in pairs of integers, starting at the beginning of \fIovector\fP, and
1820: continuing up to two-thirds of its length at the most. The first element of
1821: each pair is set to the byte offset of the first character in a substring, and
1822: the second is set to the byte offset of the first character after the end of a
1823: substring. \fBNote\fP: these values are always byte offsets, even in UTF-8
1824: mode. They are not character counts.
1825: .P
1826: The first pair of integers, \fIovector[0]\fP and \fIovector[1]\fP, identify the
1827: portion of the subject string matched by the entire pattern. The next pair is
1828: used for the first capturing subpattern, and so on. The value returned by
1829: \fBpcre_exec()\fP is one more than the highest numbered pair that has been set.
1830: For example, if two substrings have been captured, the returned value is 3. If
1831: there are no capturing subpatterns, the return value from a successful match is
1832: 1, indicating that just the first pair of offsets has been set.
1833: .P
1834: If a capturing subpattern is matched repeatedly, it is the last portion of the
1835: string that it matched that is returned.
1836: .P
1837: If the vector is too small to hold all the captured substring offsets, it is
1838: used as far as possible (up to two-thirds of its length), and the function
1839: returns a value of zero. If neither the actual string matched not any captured
1840: substrings are of interest, \fBpcre_exec()\fP may be called with \fIovector\fP
1841: passed as NULL and \fIovecsize\fP as zero. However, if the pattern contains
1842: back references and the \fIovector\fP is not big enough to remember the related
1843: substrings, PCRE has to get additional memory for use during matching. Thus it
1844: is usually advisable to supply an \fIovector\fP of reasonable size.
1845: .P
1846: There are some cases where zero is returned (indicating vector overflow) when
1847: in fact the vector is exactly the right size for the final match. For example,
1848: consider the pattern
1849: .sp
1850: (a)(?:(b)c|bd)
1851: .sp
1852: If a vector of 6 elements (allowing for only 1 captured substring) is given
1853: with subject string "abd", \fBpcre_exec()\fP will try to set the second
1854: captured string, thereby recording a vector overflow, before failing to match
1855: "c" and backing up to try the second alternative. The zero return, however,
1856: does correctly indicate that the maximum number of slots (namely 2) have been
1857: filled. In similar cases where there is temporary overflow, but the final
1858: number of used slots is actually less than the maximum, a non-zero value is
1859: returned.
1860: .P
1861: The \fBpcre_fullinfo()\fP function can be used to find out how many capturing
1862: subpatterns there are in a compiled pattern. The smallest size for
1863: \fIovector\fP that will allow for \fIn\fP captured substrings, in addition to
1864: the offsets of the substring matched by the whole pattern, is (\fIn\fP+1)*3.
1865: .P
1866: It is possible for capturing subpattern number \fIn+1\fP to match some part of
1867: the subject when subpattern \fIn\fP has not been used at all. For example, if
1868: the string "abc" is matched against the pattern (a|(z))(bc) the return from the
1869: function is 4, and subpatterns 1 and 3 are matched, but 2 is not. When this
1870: happens, both values in the offset pairs corresponding to unused subpatterns
1871: are set to -1.
1872: .P
1873: Offset values that correspond to unused subpatterns at the end of the
1874: expression are also set to -1. For example, if the string "abc" is matched
1875: against the pattern (abc)(x(yz)?)? subpatterns 2 and 3 are not matched. The
1876: return from the function is 2, because the highest used capturing subpattern
1877: number is 1, and the offsets for for the second and third capturing subpatterns
1878: (assuming the vector is large enough, of course) are set to -1.
1879: .P
1880: \fBNote\fP: Elements in the first two-thirds of \fIovector\fP that do not
1881: correspond to capturing parentheses in the pattern are never changed. That is,
1882: if a pattern contains \fIn\fP capturing parentheses, no more than
1883: \fIovector[0]\fP to \fIovector[2n+1]\fP are set by \fBpcre_exec()\fP. The other
1884: elements (in the first two-thirds) retain whatever values they previously had.
1885: .P
1886: Some convenience functions are provided for extracting the captured substrings
1887: as separate strings. These are described below.
1888: .
1889: .
1890: .\" HTML <a name="errorlist"></a>
1891: .SS "Error return values from \fBpcre_exec()\fP"
1892: .rs
1893: .sp
1894: If \fBpcre_exec()\fP fails, it returns a negative number. The following are
1895: defined in the header file:
1896: .sp
1897: PCRE_ERROR_NOMATCH (-1)
1898: .sp
1899: The subject string did not match the pattern.
1900: .sp
1901: PCRE_ERROR_NULL (-2)
1902: .sp
1903: Either \fIcode\fP or \fIsubject\fP was passed as NULL, or \fIovector\fP was
1904: NULL and \fIovecsize\fP was not zero.
1905: .sp
1906: PCRE_ERROR_BADOPTION (-3)
1907: .sp
1908: An unrecognized bit was set in the \fIoptions\fP argument.
1909: .sp
1910: PCRE_ERROR_BADMAGIC (-4)
1911: .sp
1912: PCRE stores a 4-byte "magic number" at the start of the compiled code, to catch
1913: the case when it is passed a junk pointer and to detect when a pattern that was
1914: compiled in an environment of one endianness is run in an environment with the
1915: other endianness. This is the error that PCRE gives when the magic number is
1916: not present.
1917: .sp
1918: PCRE_ERROR_UNKNOWN_OPCODE (-5)
1919: .sp
1920: While running the pattern match, an unknown item was encountered in the
1921: compiled pattern. This error could be caused by a bug in PCRE or by overwriting
1922: of the compiled pattern.
1923: .sp
1924: PCRE_ERROR_NOMEMORY (-6)
1925: .sp
1926: If a pattern contains back references, but the \fIovector\fP that is passed to
1927: \fBpcre_exec()\fP is not big enough to remember the referenced substrings, PCRE
1928: gets a block of memory at the start of matching to use for this purpose. If the
1929: call via \fBpcre_malloc()\fP fails, this error is given. The memory is
1930: automatically freed at the end of matching.
1931: .P
1932: This error is also given if \fBpcre_stack_malloc()\fP fails in
1933: \fBpcre_exec()\fP. This can happen only when PCRE has been compiled with
1934: \fB--disable-stack-for-recursion\fP.
1935: .sp
1936: PCRE_ERROR_NOSUBSTRING (-7)
1937: .sp
1938: This error is used by the \fBpcre_copy_substring()\fP,
1939: \fBpcre_get_substring()\fP, and \fBpcre_get_substring_list()\fP functions (see
1940: below). It is never returned by \fBpcre_exec()\fP.
1941: .sp
1942: PCRE_ERROR_MATCHLIMIT (-8)
1943: .sp
1944: The backtracking limit, as specified by the \fImatch_limit\fP field in a
1945: \fBpcre_extra\fP structure (or defaulted) was reached. See the description
1946: above.
1947: .sp
1948: PCRE_ERROR_CALLOUT (-9)
1949: .sp
1950: This error is never generated by \fBpcre_exec()\fP itself. It is provided for
1951: use by callout functions that want to yield a distinctive error code. See the
1952: .\" HREF
1953: \fBpcrecallout\fP
1954: .\"
1955: documentation for details.
1956: .sp
1957: PCRE_ERROR_BADUTF8 (-10)
1958: .sp
1959: A string that contains an invalid UTF-8 byte sequence was passed as a subject,
1960: and the PCRE_NO_UTF8_CHECK option was not set. If the size of the output vector
1961: (\fIovecsize\fP) is at least 2, the byte offset to the start of the the invalid
1962: UTF-8 character is placed in the first element, and a reason code is placed in
1963: the second element. The reason codes are listed in the
1964: .\" HTML <a href="#badutf8reasons">
1965: .\" </a>
1966: following section.
1967: .\"
1968: For backward compatibility, if PCRE_PARTIAL_HARD is set and the problem is a
1969: truncated UTF-8 character at the end of the subject (reason codes 1 to 5),
1970: PCRE_ERROR_SHORTUTF8 is returned instead of PCRE_ERROR_BADUTF8.
1971: .sp
1972: PCRE_ERROR_BADUTF8_OFFSET (-11)
1973: .sp
1974: The UTF-8 byte sequence that was passed as a subject was checked and found to
1975: be valid (the PCRE_NO_UTF8_CHECK option was not set), but the value of
1976: \fIstartoffset\fP did not point to the beginning of a UTF-8 character or the
1977: end of the subject.
1978: .sp
1979: PCRE_ERROR_PARTIAL (-12)
1980: .sp
1981: The subject string did not match, but it did match partially. See the
1982: .\" HREF
1983: \fBpcrepartial\fP
1984: .\"
1985: documentation for details of partial matching.
1986: .sp
1987: PCRE_ERROR_BADPARTIAL (-13)
1988: .sp
1989: This code is no longer in use. It was formerly returned when the PCRE_PARTIAL
1990: option was used with a compiled pattern containing items that were not
1991: supported for partial matching. From release 8.00 onwards, there are no
1992: restrictions on partial matching.
1993: .sp
1994: PCRE_ERROR_INTERNAL (-14)
1995: .sp
1996: An unexpected internal error has occurred. This error could be caused by a bug
1997: in PCRE or by overwriting of the compiled pattern.
1998: .sp
1999: PCRE_ERROR_BADCOUNT (-15)
2000: .sp
2001: This error is given if the value of the \fIovecsize\fP argument is negative.
2002: .sp
2003: PCRE_ERROR_RECURSIONLIMIT (-21)
2004: .sp
2005: The internal recursion limit, as specified by the \fImatch_limit_recursion\fP
2006: field in a \fBpcre_extra\fP structure (or defaulted) was reached. See the
2007: description above.
2008: .sp
2009: PCRE_ERROR_BADNEWLINE (-23)
2010: .sp
2011: An invalid combination of PCRE_NEWLINE_\fIxxx\fP options was given.
2012: .sp
2013: PCRE_ERROR_BADOFFSET (-24)
2014: .sp
2015: The value of \fIstartoffset\fP was negative or greater than the length of the
2016: subject, that is, the value in \fIlength\fP.
2017: .sp
2018: PCRE_ERROR_SHORTUTF8 (-25)
2019: .sp
2020: This error is returned instead of PCRE_ERROR_BADUTF8 when the subject string
2021: ends with a truncated UTF-8 character and the PCRE_PARTIAL_HARD option is set.
2022: Information about the failure is returned as for PCRE_ERROR_BADUTF8. It is in
2023: fact sufficient to detect this case, but this special error code for
2024: PCRE_PARTIAL_HARD precedes the implementation of returned information; it is
2025: retained for backwards compatibility.
2026: .sp
2027: PCRE_ERROR_RECURSELOOP (-26)
2028: .sp
2029: This error is returned when \fBpcre_exec()\fP detects a recursion loop within
2030: the pattern. Specifically, it means that either the whole pattern or a
2031: subpattern has been called recursively for the second time at the same position
2032: in the subject string. Some simple patterns that might do this are detected and
2033: faulted at compile time, but more complicated cases, in particular mutual
2034: recursions between two different subpatterns, cannot be detected until run
2035: time.
2036: .sp
2037: PCRE_ERROR_JIT_STACKLIMIT (-27)
2038: .sp
2039: This error is returned when a pattern that was successfully studied using the
2040: PCRE_STUDY_JIT_COMPILE option is being matched, but the memory available for
2041: the just-in-time processing stack is not large enough. See the
2042: .\" HREF
2043: \fBpcrejit\fP
2044: .\"
2045: documentation for more details.
2046: .P
2047: Error numbers -16 to -20 and -22 are not used by \fBpcre_exec()\fP.
2048: .
2049: .
2050: .\" HTML <a name="badutf8reasons"></a>
2051: .SS "Reason codes for invalid UTF-8 strings"
2052: .rs
2053: .sp
2054: When \fBpcre_exec()\fP returns either PCRE_ERROR_BADUTF8 or
2055: PCRE_ERROR_SHORTUTF8, and the size of the output vector (\fIovecsize\fP) is at
2056: least 2, the offset of the start of the invalid UTF-8 character is placed in
2057: the first output vector element (\fIovector[0]\fP) and a reason code is placed
2058: in the second element (\fIovector[1]\fP). The reason codes are given names in
2059: the \fBpcre.h\fP header file:
2060: .sp
2061: PCRE_UTF8_ERR1
2062: PCRE_UTF8_ERR2
2063: PCRE_UTF8_ERR3
2064: PCRE_UTF8_ERR4
2065: PCRE_UTF8_ERR5
2066: .sp
2067: The string ends with a truncated UTF-8 character; the code specifies how many
2068: bytes are missing (1 to 5). Although RFC 3629 restricts UTF-8 characters to be
2069: no longer than 4 bytes, the encoding scheme (originally defined by RFC 2279)
2070: allows for up to 6 bytes, and this is checked first; hence the possibility of
2071: 4 or 5 missing bytes.
2072: .sp
2073: PCRE_UTF8_ERR6
2074: PCRE_UTF8_ERR7
2075: PCRE_UTF8_ERR8
2076: PCRE_UTF8_ERR9
2077: PCRE_UTF8_ERR10
2078: .sp
2079: The two most significant bits of the 2nd, 3rd, 4th, 5th, or 6th byte of the
2080: character do not have the binary value 0b10 (that is, either the most
2081: significant bit is 0, or the next bit is 1).
2082: .sp
2083: PCRE_UTF8_ERR11
2084: PCRE_UTF8_ERR12
2085: .sp
2086: A character that is valid by the RFC 2279 rules is either 5 or 6 bytes long;
2087: these code points are excluded by RFC 3629.
2088: .sp
2089: PCRE_UTF8_ERR13
2090: .sp
2091: A 4-byte character has a value greater than 0x10fff; these code points are
2092: excluded by RFC 3629.
2093: .sp
2094: PCRE_UTF8_ERR14
2095: .sp
2096: A 3-byte character has a value in the range 0xd800 to 0xdfff; this range of
2097: code points are reserved by RFC 3629 for use with UTF-16, and so are excluded
2098: from UTF-8.
2099: .sp
2100: PCRE_UTF8_ERR15
2101: PCRE_UTF8_ERR16
2102: PCRE_UTF8_ERR17
2103: PCRE_UTF8_ERR18
2104: PCRE_UTF8_ERR19
2105: .sp
2106: A 2-, 3-, 4-, 5-, or 6-byte character is "overlong", that is, it codes for a
2107: value that can be represented by fewer bytes, which is invalid. For example,
2108: the two bytes 0xc0, 0xae give the value 0x2e, whose correct coding uses just
2109: one byte.
2110: .sp
2111: PCRE_UTF8_ERR20
2112: .sp
2113: The two most significant bits of the first byte of a character have the binary
2114: value 0b10 (that is, the most significant bit is 1 and the second is 0). Such a
2115: byte can only validly occur as the second or subsequent byte of a multi-byte
2116: character.
2117: .sp
2118: PCRE_UTF8_ERR21
2119: .sp
2120: The first byte of a character has the value 0xfe or 0xff. These values can
2121: never occur in a valid UTF-8 string.
2122: .
2123: .
2124: .SH "EXTRACTING CAPTURED SUBSTRINGS BY NUMBER"
2125: .rs
2126: .sp
2127: .B int pcre_copy_substring(const char *\fIsubject\fP, int *\fIovector\fP,
2128: .ti +5n
2129: .B int \fIstringcount\fP, int \fIstringnumber\fP, char *\fIbuffer\fP,
2130: .ti +5n
2131: .B int \fIbuffersize\fP);
2132: .PP
2133: .B int pcre_get_substring(const char *\fIsubject\fP, int *\fIovector\fP,
2134: .ti +5n
2135: .B int \fIstringcount\fP, int \fIstringnumber\fP,
2136: .ti +5n
2137: .B const char **\fIstringptr\fP);
2138: .PP
2139: .B int pcre_get_substring_list(const char *\fIsubject\fP,
2140: .ti +5n
2141: .B int *\fIovector\fP, int \fIstringcount\fP, "const char ***\fIlistptr\fP);"
2142: .PP
2143: Captured substrings can be accessed directly by using the offsets returned by
2144: \fBpcre_exec()\fP in \fIovector\fP. For convenience, the functions
2145: \fBpcre_copy_substring()\fP, \fBpcre_get_substring()\fP, and
2146: \fBpcre_get_substring_list()\fP are provided for extracting captured substrings
2147: as new, separate, zero-terminated strings. These functions identify substrings
2148: by number. The next section describes functions for extracting named
2149: substrings.
2150: .P
2151: A substring that contains a binary zero is correctly extracted and has a
2152: further zero added on the end, but the result is not, of course, a C string.
2153: However, you can process such a string by referring to the length that is
2154: returned by \fBpcre_copy_substring()\fP and \fBpcre_get_substring()\fP.
2155: Unfortunately, the interface to \fBpcre_get_substring_list()\fP is not adequate
2156: for handling strings containing binary zeros, because the end of the final
2157: string is not independently indicated.
2158: .P
2159: The first three arguments are the same for all three of these functions:
2160: \fIsubject\fP is the subject string that has just been successfully matched,
2161: \fIovector\fP is a pointer to the vector of integer offsets that was passed to
2162: \fBpcre_exec()\fP, and \fIstringcount\fP is the number of substrings that were
2163: captured by the match, including the substring that matched the entire regular
2164: expression. This is the value returned by \fBpcre_exec()\fP if it is greater
2165: than zero. If \fBpcre_exec()\fP returned zero, indicating that it ran out of
2166: space in \fIovector\fP, the value passed as \fIstringcount\fP should be the
2167: number of elements in the vector divided by three.
2168: .P
2169: The functions \fBpcre_copy_substring()\fP and \fBpcre_get_substring()\fP
2170: extract a single substring, whose number is given as \fIstringnumber\fP. A
2171: value of zero extracts the substring that matched the entire pattern, whereas
2172: higher values extract the captured substrings. For \fBpcre_copy_substring()\fP,
2173: the string is placed in \fIbuffer\fP, whose length is given by
2174: \fIbuffersize\fP, while for \fBpcre_get_substring()\fP a new block of memory is
2175: obtained via \fBpcre_malloc\fP, and its address is returned via
2176: \fIstringptr\fP. The yield of the function is the length of the string, not
2177: including the terminating zero, or one of these error codes:
2178: .sp
2179: PCRE_ERROR_NOMEMORY (-6)
2180: .sp
2181: The buffer was too small for \fBpcre_copy_substring()\fP, or the attempt to get
2182: memory failed for \fBpcre_get_substring()\fP.
2183: .sp
2184: PCRE_ERROR_NOSUBSTRING (-7)
2185: .sp
2186: There is no substring whose number is \fIstringnumber\fP.
2187: .P
2188: The \fBpcre_get_substring_list()\fP function extracts all available substrings
2189: and builds a list of pointers to them. All this is done in a single block of
2190: memory that is obtained via \fBpcre_malloc\fP. The address of the memory block
2191: is returned via \fIlistptr\fP, which is also the start of the list of string
2192: pointers. The end of the list is marked by a NULL pointer. The yield of the
2193: function is zero if all went well, or the error code
2194: .sp
2195: PCRE_ERROR_NOMEMORY (-6)
2196: .sp
2197: if the attempt to get the memory block failed.
2198: .P
2199: When any of these functions encounter a substring that is unset, which can
2200: happen when capturing subpattern number \fIn+1\fP matches some part of the
2201: subject, but subpattern \fIn\fP has not been used at all, they return an empty
2202: string. This can be distinguished from a genuine zero-length substring by
2203: inspecting the appropriate offset in \fIovector\fP, which is negative for unset
2204: substrings.
2205: .P
2206: The two convenience functions \fBpcre_free_substring()\fP and
2207: \fBpcre_free_substring_list()\fP can be used to free the memory returned by
2208: a previous call of \fBpcre_get_substring()\fP or
2209: \fBpcre_get_substring_list()\fP, respectively. They do nothing more than call
2210: the function pointed to by \fBpcre_free\fP, which of course could be called
2211: directly from a C program. However, PCRE is used in some situations where it is
2212: linked via a special interface to another programming language that cannot use
2213: \fBpcre_free\fP directly; it is for these cases that the functions are
2214: provided.
2215: .
2216: .
2217: .SH "EXTRACTING CAPTURED SUBSTRINGS BY NAME"
2218: .rs
2219: .sp
2220: .B int pcre_get_stringnumber(const pcre *\fIcode\fP,
2221: .ti +5n
2222: .B const char *\fIname\fP);
2223: .PP
2224: .B int pcre_copy_named_substring(const pcre *\fIcode\fP,
2225: .ti +5n
2226: .B const char *\fIsubject\fP, int *\fIovector\fP,
2227: .ti +5n
2228: .B int \fIstringcount\fP, const char *\fIstringname\fP,
2229: .ti +5n
2230: .B char *\fIbuffer\fP, int \fIbuffersize\fP);
2231: .PP
2232: .B int pcre_get_named_substring(const pcre *\fIcode\fP,
2233: .ti +5n
2234: .B const char *\fIsubject\fP, int *\fIovector\fP,
2235: .ti +5n
2236: .B int \fIstringcount\fP, const char *\fIstringname\fP,
2237: .ti +5n
2238: .B const char **\fIstringptr\fP);
2239: .PP
2240: To extract a substring by name, you first have to find associated number.
2241: For example, for this pattern
2242: .sp
2243: (a+)b(?<xxx>\ed+)...
2244: .sp
2245: the number of the subpattern called "xxx" is 2. If the name is known to be
2246: unique (PCRE_DUPNAMES was not set), you can find the number from the name by
2247: calling \fBpcre_get_stringnumber()\fP. The first argument is the compiled
2248: pattern, and the second is the name. The yield of the function is the
2249: subpattern number, or PCRE_ERROR_NOSUBSTRING (-7) if there is no subpattern of
2250: that name.
2251: .P
2252: Given the number, you can extract the substring directly, or use one of the
2253: functions described in the previous section. For convenience, there are also
2254: two functions that do the whole job.
2255: .P
2256: Most of the arguments of \fBpcre_copy_named_substring()\fP and
2257: \fBpcre_get_named_substring()\fP are the same as those for the similarly named
2258: functions that extract by number. As these are described in the previous
2259: section, they are not re-described here. There are just two differences:
2260: .P
2261: First, instead of a substring number, a substring name is given. Second, there
2262: is an extra argument, given at the start, which is a pointer to the compiled
2263: pattern. This is needed in order to gain access to the name-to-number
2264: translation table.
2265: .P
2266: These functions call \fBpcre_get_stringnumber()\fP, and if it succeeds, they
2267: then call \fBpcre_copy_substring()\fP or \fBpcre_get_substring()\fP, as
2268: appropriate. \fBNOTE:\fP If PCRE_DUPNAMES is set and there are duplicate names,
2269: the behaviour may not be what you want (see the next section).
2270: .P
2271: \fBWarning:\fP If the pattern uses the (?| feature to set up multiple
2272: subpatterns with the same number, as described in the
2273: .\" HTML <a href="pcrepattern.html#dupsubpatternnumber">
2274: .\" </a>
2275: section on duplicate subpattern numbers
2276: .\"
2277: in the
2278: .\" HREF
2279: \fBpcrepattern\fP
2280: .\"
2281: page, you cannot use names to distinguish the different subpatterns, because
2282: names are not included in the compiled code. The matching process uses only
2283: numbers. For this reason, the use of different names for subpatterns of the
2284: same number causes an error at compile time.
2285: .
2286: .
2287: .SH "DUPLICATE SUBPATTERN NAMES"
2288: .rs
2289: .sp
2290: .B int pcre_get_stringtable_entries(const pcre *\fIcode\fP,
2291: .ti +5n
2292: .B const char *\fIname\fP, char **\fIfirst\fP, char **\fIlast\fP);
2293: .PP
2294: When a pattern is compiled with the PCRE_DUPNAMES option, names for subpatterns
2295: are not required to be unique. (Duplicate names are always allowed for
2296: subpatterns with the same number, created by using the (?| feature. Indeed, if
2297: such subpatterns are named, they are required to use the same names.)
2298: .P
2299: Normally, patterns with duplicate names are such that in any one match, only
2300: one of the named subpatterns participates. An example is shown in the
2301: .\" HREF
2302: \fBpcrepattern\fP
2303: .\"
2304: documentation.
2305: .P
2306: When duplicates are present, \fBpcre_copy_named_substring()\fP and
2307: \fBpcre_get_named_substring()\fP return the first substring corresponding to
2308: the given name that is set. If none are set, PCRE_ERROR_NOSUBSTRING (-7) is
2309: returned; no data is returned. The \fBpcre_get_stringnumber()\fP function
2310: returns one of the numbers that are associated with the name, but it is not
2311: defined which it is.
2312: .P
2313: If you want to get full details of all captured substrings for a given name,
2314: you must use the \fBpcre_get_stringtable_entries()\fP function. The first
2315: argument is the compiled pattern, and the second is the name. The third and
2316: fourth are pointers to variables which are updated by the function. After it
2317: has run, they point to the first and last entries in the name-to-number table
2318: for the given name. The function itself returns the length of each entry, or
2319: PCRE_ERROR_NOSUBSTRING (-7) if there are none. The format of the table is
2320: described above in the section entitled \fIInformation about a pattern\fP
2321: .\" HTML <a href="#infoaboutpattern">
2322: .\" </a>
2323: above.
2324: .\"
2325: Given all the relevant entries for the name, you can extract each of their
2326: numbers, and hence the captured data, if any.
2327: .
2328: .
2329: .SH "FINDING ALL POSSIBLE MATCHES"
2330: .rs
2331: .sp
2332: The traditional matching function uses a similar algorithm to Perl, which stops
2333: when it finds the first match, starting at a given point in the subject. If you
2334: want to find all possible matches, or the longest possible match, consider
2335: using the alternative matching function (see below) instead. If you cannot use
2336: the alternative function, but still need to find all possible matches, you
2337: can kludge it up by making use of the callout facility, which is described in
2338: the
2339: .\" HREF
2340: \fBpcrecallout\fP
2341: .\"
2342: documentation.
2343: .P
2344: What you have to do is to insert a callout right at the end of the pattern.
2345: When your callout function is called, extract and save the current matched
2346: substring. Then return 1, which forces \fBpcre_exec()\fP to backtrack and try
2347: other alternatives. Ultimately, when it runs out of matches, \fBpcre_exec()\fP
2348: will yield PCRE_ERROR_NOMATCH.
2349: .
2350: .
2351: .\" HTML <a name="dfamatch"></a>
2352: .SH "MATCHING A PATTERN: THE ALTERNATIVE FUNCTION"
2353: .rs
2354: .sp
2355: .B int pcre_dfa_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
2356: .ti +5n
2357: .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
2358: .ti +5n
2359: .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP,
2360: .ti +5n
2361: .B int *\fIworkspace\fP, int \fIwscount\fP);
2362: .P
2363: The function \fBpcre_dfa_exec()\fP is called to match a subject string against
2364: a compiled pattern, using a matching algorithm that scans the subject string
2365: just once, and does not backtrack. This has different characteristics to the
2366: normal algorithm, and is not compatible with Perl. Some of the features of PCRE
2367: patterns are not supported. Nevertheless, there are times when this kind of
2368: matching can be useful. For a discussion of the two matching algorithms, and a
2369: list of features that \fBpcre_dfa_exec()\fP does not support, see the
2370: .\" HREF
2371: \fBpcrematching\fP
2372: .\"
2373: documentation.
2374: .P
2375: The arguments for the \fBpcre_dfa_exec()\fP function are the same as for
2376: \fBpcre_exec()\fP, plus two extras. The \fIovector\fP argument is used in a
2377: different way, and this is described below. The other common arguments are used
2378: in the same way as for \fBpcre_exec()\fP, so their description is not repeated
2379: here.
2380: .P
2381: The two additional arguments provide workspace for the function. The workspace
2382: vector should contain at least 20 elements. It is used for keeping track of
2383: multiple paths through the pattern tree. More workspace will be needed for
2384: patterns and subjects where there are a lot of potential matches.
2385: .P
2386: Here is an example of a simple call to \fBpcre_dfa_exec()\fP:
2387: .sp
2388: int rc;
2389: int ovector[10];
2390: int wspace[20];
2391: rc = pcre_dfa_exec(
2392: re, /* result of pcre_compile() */
2393: NULL, /* we didn't study the pattern */
2394: "some string", /* the subject string */
2395: 11, /* the length of the subject string */
2396: 0, /* start at offset 0 in the subject */
2397: 0, /* default options */
2398: ovector, /* vector of integers for substring information */
2399: 10, /* number of elements (NOT size in bytes) */
2400: wspace, /* working space vector */
2401: 20); /* number of elements (NOT size in bytes) */
2402: .
2403: .SS "Option bits for \fBpcre_dfa_exec()\fP"
2404: .rs
2405: .sp
2406: The unused bits of the \fIoptions\fP argument for \fBpcre_dfa_exec()\fP must be
2407: zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_\fIxxx\fP,
2408: PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NOTEMPTY_ATSTART,
2409: PCRE_NO_UTF8_CHECK, PCRE_BSR_ANYCRLF, PCRE_BSR_UNICODE, PCRE_NO_START_OPTIMIZE,
2410: PCRE_PARTIAL_HARD, PCRE_PARTIAL_SOFT, PCRE_DFA_SHORTEST, and PCRE_DFA_RESTART.
2411: All but the last four of these are exactly the same as for \fBpcre_exec()\fP,
2412: so their description is not repeated here.
2413: .sp
2414: PCRE_PARTIAL_HARD
2415: PCRE_PARTIAL_SOFT
2416: .sp
2417: These have the same general effect as they do for \fBpcre_exec()\fP, but the
2418: details are slightly different. When PCRE_PARTIAL_HARD is set for
2419: \fBpcre_dfa_exec()\fP, it returns PCRE_ERROR_PARTIAL if the end of the subject
2420: is reached and there is still at least one matching possibility that requires
2421: additional characters. This happens even if some complete matches have also
2422: been found. When PCRE_PARTIAL_SOFT is set, the return code PCRE_ERROR_NOMATCH
2423: is converted into PCRE_ERROR_PARTIAL if the end of the subject is reached,
2424: there have been no complete matches, but there is still at least one matching
2425: possibility. The portion of the string that was inspected when the longest
2426: partial match was found is set as the first matching string in both cases.
2427: There is a more detailed discussion of partial and multi-segment matching, with
2428: examples, in the
2429: .\" HREF
2430: \fBpcrepartial\fP
2431: .\"
2432: documentation.
2433: .sp
2434: PCRE_DFA_SHORTEST
2435: .sp
2436: Setting the PCRE_DFA_SHORTEST option causes the matching algorithm to stop as
2437: soon as it has found one match. Because of the way the alternative algorithm
2438: works, this is necessarily the shortest possible match at the first possible
2439: matching point in the subject string.
2440: .sp
2441: PCRE_DFA_RESTART
2442: .sp
2443: When \fBpcre_dfa_exec()\fP returns a partial match, it is possible to call it
2444: again, with additional subject characters, and have it continue with the same
2445: match. The PCRE_DFA_RESTART option requests this action; when it is set, the
2446: \fIworkspace\fP and \fIwscount\fP options must reference the same vector as
2447: before because data about the match so far is left in them after a partial
2448: match. There is more discussion of this facility in the
2449: .\" HREF
2450: \fBpcrepartial\fP
2451: .\"
2452: documentation.
2453: .
2454: .
2455: .SS "Successful returns from \fBpcre_dfa_exec()\fP"
2456: .rs
2457: .sp
2458: When \fBpcre_dfa_exec()\fP succeeds, it may have matched more than one
2459: substring in the subject. Note, however, that all the matches from one run of
2460: the function start at the same point in the subject. The shorter matches are
2461: all initial substrings of the longer matches. For example, if the pattern
2462: .sp
2463: <.*>
2464: .sp
2465: is matched against the string
2466: .sp
2467: This is <something> <something else> <something further> no more
2468: .sp
2469: the three matched strings are
2470: .sp
2471: <something>
2472: <something> <something else>
2473: <something> <something else> <something further>
2474: .sp
2475: On success, the yield of the function is a number greater than zero, which is
2476: the number of matched substrings. The substrings themselves are returned in
2477: \fIovector\fP. Each string uses two elements; the first is the offset to the
2478: start, and the second is the offset to the end. In fact, all the strings have
2479: the same start offset. (Space could have been saved by giving this only once,
2480: but it was decided to retain some compatibility with the way \fBpcre_exec()\fP
2481: returns data, even though the meaning of the strings is different.)
2482: .P
2483: The strings are returned in reverse order of length; that is, the longest
2484: matching string is given first. If there were too many matches to fit into
2485: \fIovector\fP, the yield of the function is zero, and the vector is filled with
2486: the longest matches. Unlike \fBpcre_exec()\fP, \fBpcre_dfa_exec()\fP can use
2487: the entire \fIovector\fP for returning matched strings.
2488: .
2489: .
2490: .SS "Error returns from \fBpcre_dfa_exec()\fP"
2491: .rs
2492: .sp
2493: The \fBpcre_dfa_exec()\fP function returns a negative number when it fails.
2494: Many of the errors are the same as for \fBpcre_exec()\fP, and these are
2495: described
2496: .\" HTML <a href="#errorlist">
2497: .\" </a>
2498: above.
2499: .\"
2500: There are in addition the following errors that are specific to
2501: \fBpcre_dfa_exec()\fP:
2502: .sp
2503: PCRE_ERROR_DFA_UITEM (-16)
2504: .sp
2505: This return is given if \fBpcre_dfa_exec()\fP encounters an item in the pattern
2506: that it does not support, for instance, the use of \eC or a back reference.
2507: .sp
2508: PCRE_ERROR_DFA_UCOND (-17)
2509: .sp
2510: This return is given if \fBpcre_dfa_exec()\fP encounters a condition item that
2511: uses a back reference for the condition, or a test for recursion in a specific
2512: group. These are not supported.
2513: .sp
2514: PCRE_ERROR_DFA_UMLIMIT (-18)
2515: .sp
2516: This return is given if \fBpcre_dfa_exec()\fP is called with an \fIextra\fP
2517: block that contains a setting of the \fImatch_limit\fP or
2518: \fImatch_limit_recursion\fP fields. This is not supported (these fields are
2519: meaningless for DFA matching).
2520: .sp
2521: PCRE_ERROR_DFA_WSSIZE (-19)
2522: .sp
2523: This return is given if \fBpcre_dfa_exec()\fP runs out of space in the
2524: \fIworkspace\fP vector.
2525: .sp
2526: PCRE_ERROR_DFA_RECURSE (-20)
2527: .sp
2528: When a recursive subpattern is processed, the matching function calls itself
2529: recursively, using private vectors for \fIovector\fP and \fIworkspace\fP. This
2530: error is given if the output vector is not large enough. This should be
2531: extremely rare, as a vector of size 1000 is used.
2532: .
2533: .
2534: .SH "SEE ALSO"
2535: .rs
2536: .sp
2537: \fBpcrebuild\fP(3), \fBpcrecallout\fP(3), \fBpcrecpp(3)\fP(3),
2538: \fBpcrematching\fP(3), \fBpcrepartial\fP(3), \fBpcreposix\fP(3),
2539: \fBpcreprecompile\fP(3), \fBpcresample\fP(3), \fBpcrestack\fP(3).
2540: .
2541: .
2542: .SH AUTHOR
2543: .rs
2544: .sp
2545: .nf
2546: Philip Hazel
2547: University Computing Service
2548: Cambridge CB2 3QH, England.
2549: .fi
2550: .
2551: .
2552: .SH REVISION
2553: .rs
2554: .sp
2555: .nf
2556: Last updated: 02 December 2011
2557: Copyright (c) 1997-2011 University of Cambridge.
2558: .fi
FreeBSD-CVSweb <freebsd-cvsweb@FreeBSD.org>