ViewVC logotype

Contents of /code/trunk/doc/pcrebuild.3

Parent Directory Parent Directory | Revision Log Revision Log

Revision 1335 - (show annotations)
Tue May 28 09:13:59 2013 UTC (8 years ago) by ph10
File size: 19745 byte(s)
Error occurred while calculating annotation data.
Final source file tidies for 8.33 release.
1 .TH PCREBUILD 3 "12 May 2013" "PCRE 8.33"
3 PCRE - Perl-compatible regular expressions
4 .
5 .
7 .rs
8 .sp
9 PCRE is distributed with a \fBconfigure\fP script that can be used to build the
10 library in Unix-like environments using the applications known as Autotools.
11 Also in the distribution are files to support building using \fBCMake\fP
12 instead of \fBconfigure\fP. The text file
13 .\" HTML <a href="README.txt">
14 .\" </a>
16 .\"
17 contains general information about building with Autotools (some of which is
18 repeated below), and also has some comments about building on various operating
19 systems. There is a lot more information about building PCRE without using
20 Autotools (including information about using \fBCMake\fP and building "by
21 hand") in the text file called
22 .\" HTML <a href="NON-AUTOTOOLS-BUILD.txt">
23 .\" </a>
25 .\"
26 You should consult this file as well as the
27 .\" HTML <a href="README.txt">
28 .\" </a>
30 .\"
31 file if you are building in a non-Unix-like environment.
32 .
33 .
35 .rs
36 .sp
37 The rest of this document describes the optional features of PCRE that can be
38 selected when the library is compiled. It assumes use of the \fBconfigure\fP
39 script, where the optional features are selected or deselected by providing
40 options to \fBconfigure\fP before running the \fBmake\fP command. However, the
41 same options can be selected in both Unix-like and non-Unix-like environments
42 using the GUI facility of \fBcmake-gui\fP if you are using \fBCMake\fP instead
43 of \fBconfigure\fP to build PCRE.
44 .P
45 If you are not using Autotools or \fBCMake\fP, option selection can be done by
46 editing the \fBconfig.h\fP file, or by passing parameter settings to the
47 compiler, as described in
48 .\" HTML <a href="NON-AUTOTOOLS-BUILD.txt">
49 .\" </a>
51 .\"
52 .P
53 The complete list of options for \fBconfigure\fP (which includes the standard
54 ones such as the selection of the installation directory) can be obtained by
55 running
56 .sp
57 ./configure --help
58 .sp
59 The following sections include descriptions of options whose names begin with
60 --enable or --disable. These settings specify changes to the defaults for the
61 \fBconfigure\fP command. Because of the way that \fBconfigure\fP works,
62 --enable and --disable always come in pairs, so the complementary option always
63 exists as well, but as it specifies the default, it is not described.
64 .
65 .
67 .rs
68 .sp
69 By default, a library called \fBlibpcre\fP is built, containing functions that
70 take string arguments contained in vectors of bytes, either as single-byte
71 characters, or interpreted as UTF-8 strings. You can also build a separate
72 library, called \fBlibpcre16\fP, in which strings are contained in vectors of
73 16-bit data units and interpreted either as single-unit characters or UTF-16
74 strings, by adding
75 .sp
76 --enable-pcre16
77 .sp
78 to the \fBconfigure\fP command. You can also build yet another separate
79 library, called \fBlibpcre32\fP, in which strings are contained in vectors of
80 32-bit data units and interpreted either as single-unit characters or UTF-32
81 strings, by adding
82 .sp
83 --enable-pcre32
84 .sp
85 to the \fBconfigure\fP command. If you do not want the 8-bit library, add
86 .sp
87 --disable-pcre8
88 .sp
89 as well. At least one of the three libraries must be built. Note that the C++
90 and POSIX wrappers are for the 8-bit library only, and that \fBpcregrep\fP is
91 an 8-bit program. None of these are built if you select only the 16-bit or
92 32-bit libraries.
93 .
94 .
96 .rs
97 .sp
98 The Autotools PCRE building process uses \fBlibtool\fP to build both shared and
99 static libraries by default. You can suppress one of these by adding one of
100 .sp
101 --disable-shared
102 --disable-static
103 .sp
104 to the \fBconfigure\fP command, as required.
105 .
106 .
107 .SH "C++ SUPPORT"
108 .rs
109 .sp
110 By default, if the 8-bit library is being built, the \fBconfigure\fP script
111 will search for a C++ compiler and C++ header files. If it finds them, it
112 automatically builds the C++ wrapper library (which supports only 8-bit
113 strings). You can disable this by adding
114 .sp
115 --disable-cpp
116 .sp
117 to the \fBconfigure\fP command.
118 .
119 .
120 .SH "UTF-8, UTF-16 AND UTF-32 SUPPORT"
121 .rs
122 .sp
123 To build PCRE with support for UTF Unicode character strings, add
124 .sp
125 --enable-utf
126 .sp
127 to the \fBconfigure\fP command. This setting applies to all three libraries,
128 adding support for UTF-8 to the 8-bit library, support for UTF-16 to the 16-bit
129 library, and support for UTF-32 to the to the 32-bit library. There are no
130 separate options for enabling UTF-8, UTF-16 and UTF-32 independently because
131 that would allow ridiculous settings such as requesting UTF-16 support while
132 building only the 8-bit library. It is not possible to build one library with
133 UTF support and another without in the same configuration. (For backwards
134 compatibility, --enable-utf8 is a synonym of --enable-utf.)
135 .P
136 Of itself, this setting does not make PCRE treat strings as UTF-8, UTF-16 or
137 UTF-32. As well as compiling PCRE with this option, you also have have to set
138 the PCRE_UTF8, PCRE_UTF16 or PCRE_UTF32 option (as appropriate) when you call
139 one of the pattern compiling functions.
140 .P
141 If you set --enable-utf when compiling in an EBCDIC environment, PCRE expects
142 its input to be either ASCII or UTF-8 (depending on the run-time option). It is
143 not possible to support both EBCDIC and UTF-8 codes in the same version of the
144 library. Consequently, --enable-utf and --enable-ebcdic are mutually
145 exclusive.
146 .
147 .
149 .rs
150 .sp
151 UTF support allows the libraries to process character codepoints up to 0x10ffff
152 in the strings that they handle. On its own, however, it does not provide any
153 facilities for accessing the properties of such characters. If you want to be
154 able to use the pattern escapes \eP, \ep, and \eX, which refer to Unicode
155 character properties, you must add
156 .sp
157 --enable-unicode-properties
158 .sp
159 to the \fBconfigure\fP command. This implies UTF support, even if you have
160 not explicitly requested it.
161 .P
162 Including Unicode property support adds around 30K of tables to the PCRE
163 library. Only the general category properties such as \fILu\fP and \fINd\fP are
164 supported. Details are given in the
165 .\" HREF
166 \fBpcrepattern\fP
167 .\"
168 documentation.
169 .
170 .
172 .rs
173 .sp
174 Just-in-time compiler support is included in the build by specifying
175 .sp
176 --enable-jit
177 .sp
178 This support is available only for certain hardware architectures. If this
179 option is set for an unsupported architecture, a compile time error occurs.
180 See the
181 .\" HREF
182 \fBpcrejit\fP
183 .\"
184 documentation for a discussion of JIT usage. When JIT support is enabled,
185 pcregrep automatically makes use of it, unless you add
186 .sp
187 --disable-pcregrep-jit
188 .sp
189 to the "configure" command.
190 .
191 .
193 .rs
194 .sp
195 By default, PCRE interprets the linefeed (LF) character as indicating the end
196 of a line. This is the normal newline character on Unix-like systems. You can
197 compile PCRE to use carriage return (CR) instead, by adding
198 .sp
199 --enable-newline-is-cr
200 .sp
201 to the \fBconfigure\fP command. There is also a --enable-newline-is-lf option,
202 which explicitly specifies linefeed as the newline character.
203 .sp
204 Alternatively, you can specify that line endings are to be indicated by the two
205 character sequence CRLF. If you want this, add
206 .sp
207 --enable-newline-is-crlf
208 .sp
209 to the \fBconfigure\fP command. There is a fourth option, specified by
210 .sp
211 --enable-newline-is-anycrlf
212 .sp
213 which causes PCRE to recognize any of the three sequences CR, LF, or CRLF as
214 indicating a line ending. Finally, a fifth option, specified by
215 .sp
216 --enable-newline-is-any
217 .sp
218 causes PCRE to recognize any Unicode newline sequence.
219 .P
220 Whatever line ending convention is selected when PCRE is built can be
221 overridden when the library functions are called. At build time it is
222 conventional to use the standard for your operating system.
223 .
224 .
226 .rs
227 .sp
228 By default, the sequence \eR in a pattern matches any Unicode newline sequence,
229 whatever has been selected as the line ending sequence. If you specify
230 .sp
231 --enable-bsr-anycrlf
232 .sp
233 the default is changed so that \eR matches only CR, LF, or CRLF. Whatever is
234 selected when PCRE is built can be overridden when the library functions are
235 called.
236 .
237 .
239 .rs
240 .sp
241 When the 8-bit library is called through the POSIX interface (see the
242 .\" HREF
243 \fBpcreposix\fP
244 .\"
245 documentation), additional working storage is required for holding the pointers
246 to capturing substrings, because PCRE requires three integers per substring,
247 whereas the POSIX interface provides only two. If the number of expected
248 substrings is small, the wrapper function uses space on the stack, because this
249 is faster than using \fBmalloc()\fP for each call. The default threshold above
250 which the stack is no longer used is 10; it can be changed by adding a setting
251 such as
252 .sp
253 --with-posix-malloc-threshold=20
254 .sp
255 to the \fBconfigure\fP command.
256 .
257 .
259 .rs
260 .sp
261 Within a compiled pattern, offset values are used to point from one part to
262 another (for example, from an opening parenthesis to an alternation
263 metacharacter). By default, in the 8-bit and 16-bit libraries, two-byte values
264 are used for these offsets, leading to a maximum size for a compiled pattern of
265 around 64K. This is sufficient to handle all but the most gigantic patterns.
266 Nevertheless, some people do want to process truly enormous patterns, so it is
267 possible to compile PCRE to use three-byte or four-byte offsets by adding a
268 setting such as
269 .sp
270 --with-link-size=3
271 .sp
272 to the \fBconfigure\fP command. The value given must be 2, 3, or 4. For the
273 16-bit library, a value of 3 is rounded up to 4. In these libraries, using
274 longer offsets slows down the operation of PCRE because it has to load
275 additional data when handling them. For the 32-bit library the value is always
276 4 and cannot be overridden; the value of --with-link-size is ignored.
277 .
278 .
280 .rs
281 .sp
282 When matching with the \fBpcre_exec()\fP function, PCRE implements backtracking
283 by making recursive calls to an internal function called \fBmatch()\fP. In
284 environments where the size of the stack is limited, this can severely limit
285 PCRE's operation. (The Unix environment does not usually suffer from this
286 problem, but it may sometimes be necessary to increase the maximum stack size.
287 There is a discussion in the
288 .\" HREF
289 \fBpcrestack\fP
290 .\"
291 documentation.) An alternative approach to recursion that uses memory from the
292 heap to remember data, instead of using recursive function calls, has been
293 implemented to work round the problem of limited stack size. If you want to
294 build a version of PCRE that works this way, add
295 .sp
296 --disable-stack-for-recursion
297 .sp
298 to the \fBconfigure\fP command. With this configuration, PCRE will use the
299 \fBpcre_stack_malloc\fP and \fBpcre_stack_free\fP variables to call memory
300 management functions. By default these point to \fBmalloc()\fP and
301 \fBfree()\fP, but you can replace the pointers so that your own functions are
302 used instead.
303 .P
304 Separate functions are provided rather than using \fBpcre_malloc\fP and
305 \fBpcre_free\fP because the usage is very predictable: the block sizes
306 requested are always the same, and the blocks are always freed in reverse
307 order. A calling program might be able to implement optimized functions that
308 perform better than \fBmalloc()\fP and \fBfree()\fP. PCRE runs noticeably more
309 slowly when built in this way. This option affects only the \fBpcre_exec()\fP
310 function; it is not relevant for \fBpcre_dfa_exec()\fP.
311 .
312 .
314 .rs
315 .sp
316 Internally, PCRE has a function called \fBmatch()\fP, which it calls repeatedly
317 (sometimes recursively) when matching a pattern with the \fBpcre_exec()\fP
318 function. By controlling the maximum number of times this function may be
319 called during a single matching operation, a limit can be placed on the
320 resources used by a single call to \fBpcre_exec()\fP. The limit can be changed
321 at run time, as described in the
322 .\" HREF
323 \fBpcreapi\fP
324 .\"
325 documentation. The default is 10 million, but this can be changed by adding a
326 setting such as
327 .sp
328 --with-match-limit=500000
329 .sp
330 to the \fBconfigure\fP command. This setting has no effect on the
331 \fBpcre_dfa_exec()\fP matching function.
332 .P
333 In some environments it is desirable to limit the depth of recursive calls of
334 \fBmatch()\fP more strictly than the total number of calls, in order to
335 restrict the maximum amount of stack (or heap, if --disable-stack-for-recursion
336 is specified) that is used. A second limit controls this; it defaults to the
337 value that is set for --with-match-limit, which imposes no additional
338 constraints. However, you can set a lower limit by adding, for example,
339 .sp
340 --with-match-limit-recursion=10000
341 .sp
342 to the \fBconfigure\fP command. This value can also be overridden at run time.
343 .
344 .
346 .rs
347 .sp
348 PCRE uses fixed tables for processing characters whose code values are less
349 than 256. By default, PCRE is built with a set of tables that are distributed
350 in the file \fIpcre_chartables.c.dist\fP. These tables are for ASCII codes
351 only. If you add
352 .sp
353 --enable-rebuild-chartables
354 .sp
355 to the \fBconfigure\fP command, the distributed tables are no longer used.
356 Instead, a program called \fBdftables\fP is compiled and run. This outputs the
357 source for new set of tables, created in the default locale of your C run-time
358 system. (This method of replacing the tables does not work if you are cross
359 compiling, because \fBdftables\fP is run on the local host. If you need to
360 create alternative tables when cross compiling, you will have to do so "by
361 hand".)
362 .
363 .
365 .rs
366 .sp
367 PCRE assumes by default that it will run in an environment where the character
368 code is ASCII (or Unicode, which is a superset of ASCII). This is the case for
369 most computer operating systems. PCRE can, however, be compiled to run in an
370 EBCDIC environment by adding
371 .sp
372 --enable-ebcdic
373 .sp
374 to the \fBconfigure\fP command. This setting implies
375 --enable-rebuild-chartables. You should only use it if you know that you are in
376 an EBCDIC environment (for example, an IBM mainframe operating system). The
377 --enable-ebcdic option is incompatible with --enable-utf.
378 .P
379 The EBCDIC character that corresponds to an ASCII LF is assumed to have the
380 value 0x15 by default. However, in some EBCDIC environments, 0x25 is used. In
381 such an environment you should use
382 .sp
383 --enable-ebcdic-nl25
384 .sp
385 as well as, or instead of, --enable-ebcdic. The EBCDIC character for CR has the
386 same value as in ASCII, namely, 0x0d. Whichever of 0x15 and 0x25 is \fInot\fP
387 chosen as LF is made to correspond to the Unicode NEL character (which, in
388 Unicode, is 0x85).
389 .P
390 The options that select newline behaviour, such as --enable-newline-is-cr,
391 and equivalent run-time options, refer to these character values in an EBCDIC
392 environment.
393 .
394 .
396 .rs
397 .sp
398 By default, \fBpcregrep\fP reads all files as plain text. You can build it so
399 that it recognizes files whose names end in \fB.gz\fP or \fB.bz2\fP, and reads
400 them with \fBlibz\fP or \fBlibbz2\fP, respectively, by adding one or both of
401 .sp
402 --enable-pcregrep-libz
403 --enable-pcregrep-libbz2
404 .sp
405 to the \fBconfigure\fP command. These options naturally require that the
406 relevant libraries are installed on your system. Configuration will fail if
407 they are not.
408 .
409 .
411 .rs
412 .sp
413 \fBpcregrep\fP uses an internal buffer to hold a "window" on the file it is
414 scanning, in order to be able to output "before" and "after" lines when it
415 finds a match. The size of the buffer is controlled by a parameter whose
416 default value is 20K. The buffer itself is three times this size, but because
417 of the way it is used for holding "before" lines, the longest line that is
418 guaranteed to be processable is the parameter size. You can change the default
419 parameter value by adding, for example,
420 .sp
421 --with-pcregrep-bufsize=50K
422 .sp
423 to the \fBconfigure\fP command. The caller of \fPpcregrep\fP can, however,
424 override this value by specifying a run-time option.
425 .
426 .
428 .rs
429 .sp
430 If you add
431 .sp
432 --enable-pcretest-libreadline
433 .sp
434 to the \fBconfigure\fP command, \fBpcretest\fP is linked with the
435 \fBlibreadline\fP library, and when its input is from a terminal, it reads it
436 using the \fBreadline()\fP function. This provides line-editing and history
437 facilities. Note that \fBlibreadline\fP is GPL-licensed, so if you distribute a
438 binary of \fBpcretest\fP linked in this way, there may be licensing issues.
439 .P
440 Setting this option causes the \fB-lreadline\fP option to be added to the
441 \fBpcretest\fP build. In many operating environments with a sytem-installed
442 \fBlibreadline\fP this is sufficient. However, in some environments (e.g.
443 if an unmodified distribution version of readline is in use), some extra
444 configuration may be necessary. The INSTALL file for \fBlibreadline\fP says
445 this:
446 .sp
447 "Readline uses the termcap functions, but does not link with the
448 termcap or curses library itself, allowing applications which link
449 with readline the to choose an appropriate library."
450 .sp
451 If your environment has not been set up so that an appropriate library is
452 automatically included, you may need to add something like
453 .sp
454 LIBS="-ncurses"
455 .sp
456 immediately before the \fBconfigure\fP command.
457 .
458 .
460 .rs
461 .sp
462 By adding the
463 .sp
464 --enable-valgrind
465 .sp
466 option to to the \fBconfigure\fP command, PCRE will use valgrind annotations
467 to mark certain memory regions as unaddressable. This allows it to detect
468 invalid memory accesses, and is mostly useful for debugging PCRE itself.
469 .
470 .
472 .rs
473 .sp
474 If your C compiler is gcc, you can build a version of PCRE that can generate a
475 code coverage report for its test suite. To enable this, you must install
476 \fBlcov\fP version 1.6 or above. Then specify
477 .sp
478 --enable-coverage
479 .sp
480 to the \fBconfigure\fP command and build PCRE in the usual way.
481 .P
482 Note that using \fBccache\fP (a caching C compiler) is incompatible with code
483 coverage reporting. If you have configured \fBccache\fP to run automatically
484 on your system, you must set the environment variable
485 .sp
487 .sp
488 before running \fBmake\fP to build PCRE, so that \fBccache\fP is not used.
489 .P
490 When --enable-coverage is used, the following addition targets are added to the
491 \fIMakefile\fP:
492 .sp
493 make coverage
494 .sp
495 This creates a fresh coverage report for the PCRE test suite. It is equivalent
496 to running "make coverage-reset", "make coverage-baseline", "make check", and
497 then "make coverage-report".
498 .sp
499 make coverage-reset
500 .sp
501 This zeroes the coverage counters, but does nothing else.
502 .sp
503 make coverage-baseline
504 .sp
505 This captures baseline coverage information.
506 .sp
507 make coverage-report
508 .sp
509 This creates the coverage report.
510 .sp
511 make coverage-clean-report
512 .sp
513 This removes the generated coverage report without cleaning the coverage data
514 itself.
515 .sp
516 make coverage-clean-data
517 .sp
518 This removes the captured coverage data without removing the coverage files
519 created at compile time (*.gcno).
520 .sp
521 make coverage-clean
522 .sp
523 This cleans all coverage data including the generated coverage report. For more
524 information about code coverage, see the \fBgcov\fP and \fBlcov\fP
525 documentation.
526 .
527 .
528 .SH "SEE ALSO"
529 .rs
530 .sp
531 \fBpcreapi\fP(3), \fBpcre16\fP, \fBpcre32\fP, \fBpcre_config\fP(3).
532 .
533 .
535 .rs
536 .sp
537 .nf
538 Philip Hazel
539 University Computing Service
540 Cambridge CB2 3QH, England.
541 .fi
542 .
543 .
545 .rs
546 .sp
547 .nf
548 Last updated: 12 May 2013
549 Copyright (c) 1997-2013 University of Cambridge.
550 .fi


Name Value
svn:eol-style native
svn:keywords "Author Date Id Revision Url"

  ViewVC Help
Powered by ViewVC 1.1.5