embedaddon/pcre/doc/html/pcrebuild.html - view

File: [ELWIX - Embedded LightWeight unIX -] / embedaddon / pcre / doc / html / pcrebuild.html
Revision 1.1.1.4 (vendor branch): download - view: text, annotated - select for diffs - revision graph
Mon Jul 22 08:25:57 2013 UTC (11 years, 8 months ago) by misho
Branches: pcre, MAIN
CVS tags: v8_34, v8_33, HEAD

8.33

1: <html> 2: <head> 3: <title>pcrebuild specification</title> 4: </head> 5: <body bgcolor="#FFFFFF" text="#00005A" link="#0066FF" alink="#3399FF" vlink="#2222BB"> 6: <h1>pcrebuild man page</h1> 7:  8: Return to the <a href="index.html">PCRE index page</a>. 9:  10:  11: This page is part of the PCRE HTML documentation. It was generated automatically 12: from the original man page. If there is any nonsense in it, please consult the 13: man page, in case the conversion went wrong. 14:   15: <ul> 16: <li><a name="TOC1" href="#SEC1">BUILDING PCRE</a> 17: <li><a name="TOC2" href="#SEC2">PCRE BUILD-TIME OPTIONS</a> 18: <li><a name="TOC3" href="#SEC3">BUILDING 8-BIT, 16-BIT AND 32-BIT LIBRARIES</a> 19: <li><a name="TOC4" href="#SEC4">BUILDING SHARED AND STATIC LIBRARIES</a> 20: <li><a name="TOC5" href="#SEC5">C++ SUPPORT</a> 21: <li><a name="TOC6" href="#SEC6">UTF-8, UTF-16 AND UTF-32 SUPPORT</a> 22: <li><a name="TOC7" href="#SEC7">UNICODE CHARACTER PROPERTY SUPPORT</a> 23: <li><a name="TOC8" href="#SEC8">JUST-IN-TIME COMPILER SUPPORT</a> 24: <li><a name="TOC9" href="#SEC9">CODE VALUE OF NEWLINE</a> 25: <li><a name="TOC10" href="#SEC10">WHAT \R MATCHES</a> 26: <li><a name="TOC11" href="#SEC11">POSIX MALLOC USAGE</a> 27: <li><a name="TOC12" href="#SEC12">HANDLING VERY LARGE PATTERNS</a> 28: <li><a name="TOC13" href="#SEC13">AVOIDING EXCESSIVE STACK USAGE</a> 29: <li><a name="TOC14" href="#SEC14">LIMITING PCRE RESOURCE USAGE</a> 30: <li><a name="TOC15" href="#SEC15">CREATING CHARACTER TABLES AT BUILD TIME</a> 31: <li><a name="TOC16" href="#SEC16">USING EBCDIC CODE</a> 32: <li><a name="TOC17" href="#SEC17">PCREGREP OPTIONS FOR COMPRESSED FILE SUPPORT</a> 33: <li><a name="TOC18" href="#SEC18">PCREGREP BUFFER SIZE</a> 34: <li><a name="TOC19" href="#SEC19">PCRETEST OPTION FOR LIBREADLINE SUPPORT</a> 35: <li><a name="TOC20" href="#SEC20">DEBUGGING WITH VALGRIND SUPPORT</a> 36: <li><a name="TOC21" href="#SEC21">CODE COVERAGE REPORTING</a> 37: <li><a name="TOC22" href="#SEC22">SEE ALSO</a> 38: <li><a name="TOC23" href="#SEC23">AUTHOR</a> 39: <li><a name="TOC24" href="#SEC24">REVISION</a> 40: </ul> 41:  <a name="SEC1" href="#TOC1">BUILDING PCRE</a>  42:  43: PCRE is distributed with a configure script that can be used to build the 44: library in Unix-like environments using the applications known as Autotools. 45: Also in the distribution are files to support building using CMake 46: instead of configure. The text file 47: <a href="README.txt">README</a> 48: contains general information about building with Autotools (some of which is 49: repeated below), and also has some comments about building on various operating 50: systems. There is a lot more information about building PCRE without using 51: Autotools (including information about using CMake and building "by 52: hand") in the text file called 53: <a href="NON-AUTOTOOLS-BUILD.txt">NON-AUTOTOOLS-BUILD.</a> 54: You should consult this file as well as the 55: <a href="README.txt">README</a> 56: file if you are building in a non-Unix-like environment. 57:  58:  <a name="SEC2" href="#TOC1">PCRE BUILD-TIME OPTIONS</a>  59:  60: The rest of this document describes the optional features of PCRE that can be 61: selected when the library is compiled. It assumes use of the configure 62: script, where the optional features are selected or deselected by providing 63: options to configure before running the make command. However, the 64: same options can be selected in both Unix-like and non-Unix-like environments 65: using the GUI facility of cmake-gui if you are using CMake instead 66: of configure to build PCRE. 67:  68:  69: If you are not using Autotools or CMake, option selection can be done by 70: editing the config.h file, or by passing parameter settings to the 71: compiler, as described in 72: <a href="NON-AUTOTOOLS-BUILD.txt">NON-AUTOTOOLS-BUILD.</a> 73:  74:  75: The complete list of options for configure (which includes the standard 76: ones such as the selection of the installation directory) can be obtained by 77: running 78: <pre> 79: ./configure --help 80: </pre> 81: The following sections include descriptions of options whose names begin with 82: --enable or --disable. These settings specify changes to the defaults for the 83: configure command. Because of the way that configure works, 84: --enable and --disable always come in pairs, so the complementary option always 85: exists as well, but as it specifies the default, it is not described. 86:  87:  <a name="SEC3" href="#TOC1">BUILDING 8-BIT, 16-BIT AND 32-BIT LIBRARIES</a>  88:  89: By default, a library called libpcre is built, containing functions that 90: take string arguments contained in vectors of bytes, either as single-byte 91: characters, or interpreted as UTF-8 strings. You can also build a separate 92: library, called libpcre16, in which strings are contained in vectors of 93: 16-bit data units and interpreted either as single-unit characters or UTF-16 94: strings, by adding 95: <pre> 96: --enable-pcre16 97: </pre> 98: to the configure command. You can also build yet another separate 99: library, called libpcre32, in which strings are contained in vectors of 100: 32-bit data units and interpreted either as single-unit characters or UTF-32 101: strings, by adding 102: <pre> 103: --enable-pcre32 104: </pre> 105: to the configure command. If you do not want the 8-bit library, add 106: <pre> 107: --disable-pcre8 108: </pre> 109: as well. At least one of the three libraries must be built. Note that the C++ 110: and POSIX wrappers are for the 8-bit library only, and that pcregrep is 111: an 8-bit program. None of these are built if you select only the 16-bit or 112: 32-bit libraries. 113:  114:  <a name="SEC4" href="#TOC1">BUILDING SHARED AND STATIC LIBRARIES</a>  115:  116: The Autotools PCRE building process uses libtool to build both shared and 117: static libraries by default. You can suppress one of these by adding one of 118: <pre> 119: --disable-shared 120: --disable-static 121: </pre> 122: to the configure command, as required. 123:  124:  <a name="SEC5" href="#TOC1">C++ SUPPORT</a>  125:  126: By default, if the 8-bit library is being built, the configure script 127: will search for a C++ compiler and C++ header files. If it finds them, it 128: automatically builds the C++ wrapper library (which supports only 8-bit 129: strings). You can disable this by adding 130: <pre> 131: --disable-cpp 132: </pre> 133: to the configure command. 134:  135:  <a name="SEC6" href="#TOC1">UTF-8, UTF-16 AND UTF-32 SUPPORT</a>  136:  137: To build PCRE with support for UTF Unicode character strings, add 138: <pre> 139: --enable-utf 140: </pre> 141: to the configure command. This setting applies to all three libraries, 142: adding support for UTF-8 to the 8-bit library, support for UTF-16 to the 16-bit 143: library, and support for UTF-32 to the to the 32-bit library. There are no 144: separate options for enabling UTF-8, UTF-16 and UTF-32 independently because 145: that would allow ridiculous settings such as requesting UTF-16 support while 146: building only the 8-bit library. It is not possible to build one library with 147: UTF support and another without in the same configuration. (For backwards 148: compatibility, --enable-utf8 is a synonym of --enable-utf.) 149:  150:  151: Of itself, this setting does not make PCRE treat strings as UTF-8, UTF-16 or 152: UTF-32. As well as compiling PCRE with this option, you also have have to set 153: the PCRE_UTF8, PCRE_UTF16 or PCRE_UTF32 option (as appropriate) when you call 154: one of the pattern compiling functions. 155:  156:  157: If you set --enable-utf when compiling in an EBCDIC environment, PCRE expects 158: its input to be either ASCII or UTF-8 (depending on the run-time option). It is 159: not possible to support both EBCDIC and UTF-8 codes in the same version of the 160: library. Consequently, --enable-utf and --enable-ebcdic are mutually 161: exclusive. 162:  163:  <a name="SEC7" href="#TOC1">UNICODE CHARACTER PROPERTY SUPPORT</a>  164:  165: UTF support allows the libraries to process character codepoints up to 0x10ffff 166: in the strings that they handle. On its own, however, it does not provide any 167: facilities for accessing the properties of such characters. If you want to be 168: able to use the pattern escapes \P, \p, and \X, which refer to Unicode 169: character properties, you must add 170: <pre> 171: --enable-unicode-properties 172: </pre> 173: to the configure command. This implies UTF support, even if you have 174: not explicitly requested it. 175:  176:  177: Including Unicode property support adds around 30K of tables to the PCRE 178: library. Only the general category properties such as Lu and Nd are 179: supported. Details are given in the 180: <a href="pcrepattern.html">pcrepattern</a> 181: documentation. 182:  183:  <a name="SEC8" href="#TOC1">JUST-IN-TIME COMPILER SUPPORT</a>  184:  185: Just-in-time compiler support is included in the build by specifying 186: <pre> 187: --enable-jit 188: </pre> 189: This support is available only for certain hardware architectures. If this 190: option is set for an unsupported architecture, a compile time error occurs. 191: See the 192: <a href="pcrejit.html">pcrejit</a> 193: documentation for a discussion of JIT usage. When JIT support is enabled, 194: pcregrep automatically makes use of it, unless you add 195: <pre> 196: --disable-pcregrep-jit 197: </pre> 198: to the "configure" command. 199:  200:  <a name="SEC9" href="#TOC1">CODE VALUE OF NEWLINE</a>  201:  202: By default, PCRE interprets the linefeed (LF) character as indicating the end 203: of a line. This is the normal newline character on Unix-like systems. You can 204: compile PCRE to use carriage return (CR) instead, by adding 205: <pre> 206: --enable-newline-is-cr 207: </pre> 208: to the configure command. There is also a --enable-newline-is-lf option, 209: which explicitly specifies linefeed as the newline character. 210:   211:   212: Alternatively, you can specify that line endings are to be indicated by the two 213: character sequence CRLF. If you want this, add 214: <pre> 215: --enable-newline-is-crlf 216: </pre> 217: to the configure command. There is a fourth option, specified by 218: <pre> 219: --enable-newline-is-anycrlf 220: </pre> 221: which causes PCRE to recognize any of the three sequences CR, LF, or CRLF as 222: indicating a line ending. Finally, a fifth option, specified by 223: <pre> 224: --enable-newline-is-any 225: </pre> 226: causes PCRE to recognize any Unicode newline sequence. 227:  228:  229: Whatever line ending convention is selected when PCRE is built can be 230: overridden when the library functions are called. At build time it is 231: conventional to use the standard for your operating system. 232:  233:  <a name="SEC10" href="#TOC1">WHAT \R MATCHES</a>  234:  235: By default, the sequence \R in a pattern matches any Unicode newline sequence, 236: whatever has been selected as the line ending sequence. If you specify 237: <pre> 238: --enable-bsr-anycrlf 239: </pre> 240: the default is changed so that \R matches only CR, LF, or CRLF. Whatever is 241: selected when PCRE is built can be overridden when the library functions are 242: called. 243:  244:  <a name="SEC11" href="#TOC1">POSIX MALLOC USAGE</a>  245:  246: When the 8-bit library is called through the POSIX interface (see the 247: <a href="pcreposix.html">pcreposix</a> 248: documentation), additional working storage is required for holding the pointers 249: to capturing substrings, because PCRE requires three integers per substring, 250: whereas the POSIX interface provides only two. If the number of expected 251: substrings is small, the wrapper function uses space on the stack, because this 252: is faster than using malloc() for each call. The default threshold above 253: which the stack is no longer used is 10; it can be changed by adding a setting 254: such as 255: <pre> 256: --with-posix-malloc-threshold=20 257: </pre> 258: to the configure command. 259:  260:  <a name="SEC12" href="#TOC1">HANDLING VERY LARGE PATTERNS</a>  261:  262: Within a compiled pattern, offset values are used to point from one part to 263: another (for example, from an opening parenthesis to an alternation 264: metacharacter). By default, in the 8-bit and 16-bit libraries, two-byte values 265: are used for these offsets, leading to a maximum size for a compiled pattern of 266: around 64K. This is sufficient to handle all but the most gigantic patterns. 267: Nevertheless, some people do want to process truly enormous patterns, so it is 268: possible to compile PCRE to use three-byte or four-byte offsets by adding a 269: setting such as 270: <pre> 271: --with-link-size=3 272: </pre> 273: to the configure command. The value given must be 2, 3, or 4. For the 274: 16-bit library, a value of 3 is rounded up to 4. In these libraries, using 275: longer offsets slows down the operation of PCRE because it has to load 276: additional data when handling them. For the 32-bit library the value is always 277: 4 and cannot be overridden; the value of --with-link-size is ignored. 278:  279:  <a name="SEC13" href="#TOC1">AVOIDING EXCESSIVE STACK USAGE</a>  280:  281: When matching with the pcre_exec() function, PCRE implements backtracking 282: by making recursive calls to an internal function called match(). In 283: environments where the size of the stack is limited, this can severely limit 284: PCRE's operation. (The Unix environment does not usually suffer from this 285: problem, but it may sometimes be necessary to increase the maximum stack size. 286: There is a discussion in the 287: <a href="pcrestack.html">pcrestack</a> 288: documentation.) An alternative approach to recursion that uses memory from the 289: heap to remember data, instead of using recursive function calls, has been 290: implemented to work round the problem of limited stack size. If you want to 291: build a version of PCRE that works this way, add 292: <pre> 293: --disable-stack-for-recursion 294: </pre> 295: to the configure command. With this configuration, PCRE will use the 296: pcre_stack_malloc and pcre_stack_free variables to call memory 297: management functions. By default these point to malloc() and 298: free(), but you can replace the pointers so that your own functions are 299: used instead. 300:  301:  302: Separate functions are provided rather than using pcre_malloc and 303: pcre_free because the usage is very predictable: the block sizes 304: requested are always the same, and the blocks are always freed in reverse 305: order. A calling program might be able to implement optimized functions that 306: perform better than malloc() and free(). PCRE runs noticeably more 307: slowly when built in this way. This option affects only the pcre_exec() 308: function; it is not relevant for pcre_dfa_exec(). 309:  310:  <a name="SEC14" href="#TOC1">LIMITING PCRE RESOURCE USAGE</a>  311:  312: Internally, PCRE has a function called match(), which it calls repeatedly 313: (sometimes recursively) when matching a pattern with the pcre_exec() 314: function. By controlling the maximum number of times this function may be 315: called during a single matching operation, a limit can be placed on the 316: resources used by a single call to pcre_exec(). The limit can be changed 317: at run time, as described in the 318: <a href="pcreapi.html">pcreapi</a> 319: documentation. The default is 10 million, but this can be changed by adding a 320: setting such as 321: <pre> 322: --with-match-limit=500000 323: </pre> 324: to the configure command. This setting has no effect on the 325: pcre_dfa_exec() matching function. 326:  327:  328: In some environments it is desirable to limit the depth of recursive calls of 329: match() more strictly than the total number of calls, in order to 330: restrict the maximum amount of stack (or heap, if --disable-stack-for-recursion 331: is specified) that is used. A second limit controls this; it defaults to the 332: value that is set for --with-match-limit, which imposes no additional 333: constraints. However, you can set a lower limit by adding, for example, 334: <pre> 335: --with-match-limit-recursion=10000 336: </pre> 337: to the configure command. This value can also be overridden at run time. 338:  339:  <a name="SEC15" href="#TOC1">CREATING CHARACTER TABLES AT BUILD TIME</a>  340:  341: PCRE uses fixed tables for processing characters whose code values are less 342: than 256. By default, PCRE is built with a set of tables that are distributed 343: in the file pcre_chartables.c.dist. These tables are for ASCII codes 344: only. If you add 345: <pre> 346: --enable-rebuild-chartables 347: </pre> 348: to the configure command, the distributed tables are no longer used. 349: Instead, a program called dftables is compiled and run. This outputs the 350: source for new set of tables, created in the default locale of your C run-time 351: system. (This method of replacing the tables does not work if you are cross 352: compiling, because dftables is run on the local host. If you need to 353: create alternative tables when cross compiling, you will have to do so "by 354: hand".) 355:  356:  <a name="SEC16" href="#TOC1">USING EBCDIC CODE</a>  357:  358: PCRE assumes by default that it will run in an environment where the character 359: code is ASCII (or Unicode, which is a superset of ASCII). This is the case for 360: most computer operating systems. PCRE can, however, be compiled to run in an 361: EBCDIC environment by adding 362: <pre> 363: --enable-ebcdic 364: </pre> 365: to the configure command. This setting implies 366: --enable-rebuild-chartables. You should only use it if you know that you are in 367: an EBCDIC environment (for example, an IBM mainframe operating system). The 368: --enable-ebcdic option is incompatible with --enable-utf. 369:  370:  371: The EBCDIC character that corresponds to an ASCII LF is assumed to have the 372: value 0x15 by default. However, in some EBCDIC environments, 0x25 is used. In 373: such an environment you should use 374: <pre> 375: --enable-ebcdic-nl25 376: </pre> 377: as well as, or instead of, --enable-ebcdic. The EBCDIC character for CR has the 378: same value as in ASCII, namely, 0x0d. Whichever of 0x15 and 0x25 is not 379: chosen as LF is made to correspond to the Unicode NEL character (which, in 380: Unicode, is 0x85). 381:  382:  383: The options that select newline behaviour, such as --enable-newline-is-cr, 384: and equivalent run-time options, refer to these character values in an EBCDIC 385: environment. 386:  387:  <a name="SEC17" href="#TOC1">PCREGREP OPTIONS FOR COMPRESSED FILE SUPPORT</a>  388:  389: By default, pcregrep reads all files as plain text. You can build it so 390: that it recognizes files whose names end in .gz or .bz2, and reads 391: them with libz or libbz2, respectively, by adding one or both of 392: <pre> 393: --enable-pcregrep-libz 394: --enable-pcregrep-libbz2 395: </pre> 396: to the configure command. These options naturally require that the 397: relevant libraries are installed on your system. Configuration will fail if 398: they are not. 399:  400:  <a name="SEC18" href="#TOC1">PCREGREP BUFFER SIZE</a>  401:  402: pcregrep uses an internal buffer to hold a "window" on the file it is 403: scanning, in order to be able to output "before" and "after" lines when it 404: finds a match. The size of the buffer is controlled by a parameter whose 405: default value is 20K. The buffer itself is three times this size, but because 406: of the way it is used for holding "before" lines, the longest line that is 407: guaranteed to be processable is the parameter size. You can change the default 408: parameter value by adding, for example, 409: <pre> 410: --with-pcregrep-bufsize=50K 411: </pre> 412: to the configure command. The caller of \fPpcregrep\fP can, however, 413: override this value by specifying a run-time option. 414:  415:  <a name="SEC19" href="#TOC1">PCRETEST OPTION FOR LIBREADLINE SUPPORT</a>  416:  417: If you add 418: <pre> 419: --enable-pcretest-libreadline 420: </pre> 421: to the configure command, pcretest is linked with the 422: libreadline library, and when its input is from a terminal, it reads it 423: using the readline() function. This provides line-editing and history 424: facilities. Note that libreadline is GPL-licensed, so if you distribute a 425: binary of pcretest linked in this way, there may be licensing issues. 426:  427:  428: Setting this option causes the -lreadline option to be added to the 429: pcretest build. In many operating environments with a sytem-installed 430: libreadline this is sufficient. However, in some environments (e.g. 431: if an unmodified distribution version of readline is in use), some extra 432: configuration may be necessary. The INSTALL file for libreadline says 433: this: 434: <pre> 435: "Readline uses the termcap functions, but does not link with the 436: termcap or curses library itself, allowing applications which link 437: with readline the to choose an appropriate library." 438: </pre> 439: If your environment has not been set up so that an appropriate library is 440: automatically included, you may need to add something like 441: <pre> 442: LIBS="-ncurses" 443: </pre> 444: immediately before the configure command. 445:  446:  <a name="SEC20" href="#TOC1">DEBUGGING WITH VALGRIND SUPPORT</a>  447:  448: By adding the 449: <pre> 450: --enable-valgrind 451: </pre> 452: option to to the configure command, PCRE will use valgrind annotations 453: to mark certain memory regions as unaddressable. This allows it to detect 454: invalid memory accesses, and is mostly useful for debugging PCRE itself. 455:  456:  <a name="SEC21" href="#TOC1">CODE COVERAGE REPORTING</a>  457:  458: If your C compiler is gcc, you can build a version of PCRE that can generate a 459: code coverage report for its test suite. To enable this, you must install 460: lcov version 1.6 or above. Then specify 461: <pre> 462: --enable-coverage 463: </pre> 464: to the configure command and build PCRE in the usual way. 465:  466:  467: Note that using ccache (a caching C compiler) is incompatible with code 468: coverage reporting. If you have configured ccache to run automatically 469: on your system, you must set the environment variable 470: <pre> 471: CCACHE_DISABLE=1 472: </pre> 473: before running make to build PCRE, so that ccache is not used. 474:  475:  476: When --enable-coverage is used, the following addition targets are added to the 477: Makefile: 478: <pre> 479: make coverage 480: </pre> 481: This creates a fresh coverage report for the PCRE test suite. It is equivalent 482: to running "make coverage-reset", "make coverage-baseline", "make check", and 483: then "make coverage-report". 484: <pre> 485: make coverage-reset 486: </pre> 487: This zeroes the coverage counters, but does nothing else. 488: <pre> 489: make coverage-baseline 490: </pre> 491: This captures baseline coverage information. 492: <pre> 493: make coverage-report 494: </pre> 495: This creates the coverage report. 496: <pre> 497: make coverage-clean-report 498: </pre> 499: This removes the generated coverage report without cleaning the coverage data 500: itself. 501: <pre> 502: make coverage-clean-data 503: </pre> 504: This removes the captured coverage data without removing the coverage files 505: created at compile time (*.gcno). 506: <pre> 507: make coverage-clean 508: </pre> 509: This cleans all coverage data including the generated coverage report. For more 510: information about code coverage, see the gcov and lcov 511: documentation. 512:  513:  <a name="SEC22" href="#TOC1">SEE ALSO</a>  514:  515: pcreapi(3), pcre16, pcre32, pcre_config(3). 516:  517:  <a name="SEC23" href="#TOC1">AUTHOR</a>  518:  519: Philip Hazel 520:   521: University Computing Service 522:   523: Cambridge CB2 3QH, England. 524:   525:  526:  <a name="SEC24" href="#TOC1">REVISION</a>  527:  528: Last updated: 12 May 2013 529:   530: Copyright © 1997-2013 University of Cambridge. 531:   532:  533: Return to the <a href="index.html">PCRE index page</a>. 534: 