Commit | Line | Data |
9fba2765 |
1 | If you read this file _as_is_, just ignore the funny characters you\r |
2 | see. It is written in the POD format (see pod/perlpod.pod) which is\r |
3 | specially designed to be readable as is.\r |
4 | \r |
5 | =head1 NAME\r |
6 | \r |
7 | perlwin32 - Perl under Windows\r |
8 | \r |
9 | =head1 SYNOPSIS\r |
10 | \r |
11 | These are instructions for building Perl under Windows 9x/NT/2000/XP\r |
12 | on the Intel x86 and Itanium architectures.\r |
13 | \r |
14 | =head1 DESCRIPTION\r |
15 | \r |
16 | Before you start, you should glance through the README file\r |
17 | found in the top-level directory to which the Perl distribution\r |
18 | was extracted. Make sure you read and understand the terms under\r |
19 | which this software is being distributed.\r |
20 | \r |
21 | Also make sure you read L<BUGS AND CAVEATS> below for the\r |
22 | known limitations of this port.\r |
23 | \r |
24 | The INSTALL file in the perl top-level has much information that is\r |
25 | only relevant to people building Perl on Unix-like systems. In\r |
26 | particular, you can safely ignore any information that talks about\r |
27 | "Configure".\r |
28 | \r |
29 | You may also want to look at two other options for building\r |
30 | a perl that will work on Windows NT: the README.cygwin and\r |
31 | README.os2 files, each of which give a different set of rules to\r |
32 | build a Perl that will work on Win32 platforms. Those two methods\r |
33 | will probably enable you to build a more Unix-compatible perl, but\r |
34 | you will also need to download and use various other build-time and\r |
35 | run-time support software described in those files.\r |
36 | \r |
37 | This set of instructions is meant to describe a so-called "native"\r |
38 | port of Perl to Win32 platforms. This includes both 32-bit and\r |
39 | 64-bit Windows operating systems. The resulting Perl requires no\r |
40 | additional software to run (other than what came with your operating\r |
41 | system). Currently, this port is capable of using one of the\r |
42 | following compilers on the Intel x86 architecture:\r |
43 | \r |
44 | Borland C++ version 5.02 or later\r |
45 | Microsoft Visual C++ version 4.2 or later\r |
46 | Mingw32 with GCC version 2.95.2 or better\r |
47 | \r |
48 | The last of these is a high quality freeware compiler. Support\r |
49 | for it is still experimental. (Older versions of GCC are known\r |
50 | not to work.)\r |
51 | \r |
52 | This port can also be built on the Intel IA64 using:\r |
53 | \r |
54 | Microsoft Platform SDK Nov 2001 (64-bit compiler and tools)\r |
55 | \r |
56 | The MS Platform SDK can be downloaded from http://www.microsoft.com/.\r |
57 | \r |
58 | This port fully supports MakeMaker (the set of modules that\r |
59 | is used to build extensions to perl). Therefore, you should be\r |
60 | able to build and install most extensions found in the CPAN sites.\r |
61 | See L<Usage Hints for Perl on Win32> below for general hints about this.\r |
62 | \r |
63 | =head2 Setting Up Perl on Win32\r |
64 | \r |
65 | =over 4\r |
66 | \r |
67 | =item Make\r |
68 | \r |
69 | You need a "make" program to build the sources. If you are using\r |
70 | Visual C++ or the Platform SDK tools under Windows NT/2000/XP, nmake\r |
71 | will work. All other builds need dmake.\r |
72 | \r |
73 | dmake is a freely available make that has very nice macro features\r |
74 | and parallelability.\r |
75 | \r |
76 | A port of dmake for Windows is available from:\r |
77 | \r |
78 | http://www.cpan.org/authors/id/GSAR/dmake-4.1pl1-win32.zip\r |
79 | \r |
80 | (This is a fixed version of the original dmake sources obtained from\r |
81 | http://www.wticorp.com/ As of version 4.1PL1, the original\r |
82 | sources did not build as shipped and had various other problems.\r |
83 | A patch is included in the above fixed version.)\r |
84 | \r |
85 | Fetch and install dmake somewhere on your path (follow the instructions\r |
86 | in the README.NOW file).\r |
87 | \r |
88 | There exists a minor coexistence problem with dmake and Borland C++\r |
89 | compilers. Namely, if a distribution has C files named with mixed\r |
90 | case letters, they will be compiled into appropriate .obj-files named\r |
91 | with all lowercase letters, and every time dmake is invoked\r |
92 | to bring files up to date, it will try to recompile such files again.\r |
93 | For example, Tk distribution has a lot of such files, resulting in\r |
94 | needless recompiles every time dmake is invoked. To avoid this, you\r |
95 | may use the script "sync_ext.pl" after a successful build. It is\r |
96 | available in the win32 subdirectory of the Perl source distribution.\r |
97 | \r |
98 | =item Command Shell\r |
99 | \r |
100 | Use the default "cmd" shell that comes with NT. Some versions of the\r |
101 | popular 4DOS/NT shell have incompatibilities that may cause you trouble.\r |
102 | If the build fails under that shell, try building again with the cmd\r |
103 | shell.\r |
104 | \r |
105 | The nmake Makefile also has known incompatibilities with the\r |
106 | "command.com" shell that comes with Windows 9x. You will need to\r |
107 | use dmake and makefile.mk to build under Windows 9x.\r |
108 | \r |
109 | The surest way to build it is on Windows NT/2000/XP, using the cmd shell.\r |
110 | \r |
111 | Make sure the path to the build directory does not contain spaces. The\r |
112 | build usually works in this circumstance, but some tests will fail.\r |
113 | \r |
114 | =item Borland C++\r |
115 | \r |
116 | If you are using the Borland compiler, you will need dmake.\r |
117 | (The make that Borland supplies is seriously crippled and will not\r |
118 | work for MakeMaker builds.)\r |
119 | \r |
120 | See L</"Make"> above.\r |
121 | \r |
122 | =item Microsoft Visual C++\r |
123 | \r |
124 | The nmake that comes with Visual C++ will suffice for building.\r |
125 | You will need to run the VCVARS32.BAT file, usually found somewhere\r |
126 | like C:\MSDEV4.2\BIN. This will set your build environment.\r |
127 | \r |
128 | You can also use dmake to build using Visual C++; provided, however,\r |
129 | you set OSRELEASE to "microsft" (or whatever the directory name\r |
130 | under which the Visual C dmake configuration lives) in your environment\r |
131 | and edit win32/config.vc to change "make=nmake" into "make=dmake". The\r |
132 | latter step is only essential if you want to use dmake as your default\r |
133 | make for building extensions using MakeMaker.\r |
134 | \r |
135 | =item Microsoft Platform SDK 64-bit Compiler\r |
136 | \r |
137 | The nmake that comes with the Platform SDK will suffice for building\r |
138 | Perl. Make sure you are building within one of the "Build Environment"\r |
139 | shells available after you install the Platform SDK from the Start Menu.\r |
140 | \r |
141 | =item Mingw32 with GCC\r |
142 | \r |
143 | GCC-2.95.2 binaries can be downloaded from:\r |
144 | \r |
145 | ftp://ftp.xraylith.wisc.edu/pub/khan/gnu-win32/mingw32/\r |
146 | \r |
147 | You also need dmake. See L</"Make"> above on how to get it.\r |
148 | \r |
149 | The GCC-2.95.2 bundle comes with Mingw32 libraries and headers.\r |
150 | \r |
151 | Make sure you install the binaries that work with MSVCRT.DLL as indicated\r |
152 | in the README for the GCC bundle. You may need to set up a few environment\r |
153 | variables (usually ran from a batch file).\r |
154 | \r |
155 | There are a couple of problems with the version of gcc-2.95.2-msvcrt.exe\r |
156 | released 7 November 1999:\r |
157 | \r |
158 | =over\r |
159 | \r |
160 | =item *\r |
161 | \r |
162 | It left out a fix for certain command line quotes. To fix this, be sure\r |
163 | to download and install the file fixes/quote-fix-msvcrt.exe from the above\r |
164 | ftp location.\r |
165 | \r |
166 | =item *\r |
167 | \r |
168 | The definition of the fpos_t type in stdio.h may be wrong. If your\r |
169 | stdio.h has this problem, you will see an exception when running the\r |
170 | test t/lib/io_xs.t. To fix this, change the typedef for fpos_t from\r |
171 | "long" to "long long" in the file i386-mingw32msvc/include/stdio.h,\r |
172 | and rebuild.\r |
173 | \r |
174 | =back\r |
175 | \r |
176 | A potentially simpler to install (but probably soon-to-be-outdated) bundle\r |
177 | of the above package with the mentioned fixes already applied is available\r |
178 | here:\r |
179 | \r |
180 | http://downloads.ActiveState.com/pub/staff/gsar/gcc-2.95.2-msvcrt.zip\r |
181 | ftp://ftp.ActiveState.com/pub/staff/gsar/gcc-2.95.2-msvcrt.zip\r |
182 | \r |
183 | =back\r |
184 | \r |
185 | =head2 Building\r |
186 | \r |
187 | =over 4\r |
188 | \r |
189 | =item *\r |
190 | \r |
191 | Make sure you are in the "win32" subdirectory under the perl toplevel.\r |
192 | This directory contains a "Makefile" that will work with\r |
193 | versions of nmake that come with Visual C++ or the Platform SDK, and\r |
194 | a dmake "makefile.mk" that will work for all supported compilers. The\r |
195 | defaults in the dmake makefile are setup to build using the GCC compiler.\r |
196 | \r |
197 | =item *\r |
198 | \r |
199 | Edit the makefile.mk (or Makefile, if you're using nmake) and change \r |
200 | the values of INST_DRV and INST_TOP. You can also enable various\r |
201 | build flags. These are explained in the makefiles.\r |
202 | \r |
203 | You will have to make sure that CCTYPE is set correctly and that \r |
204 | CCHOME points to wherever you installed your compiler.\r |
205 | \r |
206 | The default value for CCHOME in the makefiles for Visual C++\r |
207 | may not be correct for some versions. Make sure the default exists\r |
208 | and is valid.\r |
209 | \r |
210 | If you have either the source or a library that contains des_fcrypt(),\r |
211 | enable the appropriate option in the makefile. des_fcrypt() is not\r |
212 | bundled with the distribution due to US Government restrictions\r |
213 | on the export of cryptographic software. Nevertheless, this routine\r |
214 | is part of the "libdes" library (written by Eric Young) which is widely\r |
215 | available worldwide, usually along with SSLeay ( for example, \r |
216 | ftp://ftp.funet.fi/pub/crypt/mirrors/dsi/libdes/ ). Set CRYPT_SRC to the\r |
217 | name of the file that implements des_fcrypt(). Alternatively, if\r |
218 | you have built a library that contains des_fcrypt(), you can set\r |
219 | CRYPT_LIB to point to the library name. The location above contains\r |
220 | many versions of the "libdes" library, all with slightly different\r |
221 | implementations of des_fcrypt(). Older versions have a single,\r |
222 | self-contained file (fcrypt.c) that implements crypt(), so they may be\r |
223 | easier to use. A patch against the fcrypt.c found in libdes-3.06 is\r |
224 | in des_fcrypt.patch.\r |
225 | \r |
226 | An easier alternative may be to get the pre-patched and ready-to-use\r |
227 | fcrypt.c that can be found here:\r |
228 | \r |
229 | http://downloads.ActiveState.com/pub/staff/gsar/fcrypt.c\r |
230 | ftp://ftp.ActiveState.com/pub/staff/gsar/fcrypt.c\r |
231 | \r |
232 | Perl will also build without des_fcrypt(), but the crypt() builtin will\r |
233 | fail at run time.\r |
234 | \r |
235 | Be sure to read the instructions near the top of the makefiles carefully.\r |
236 | \r |
237 | =item *\r |
238 | \r |
239 | Type "dmake" (or "nmake" if you are using that make).\r |
240 | \r |
241 | This should build everything. Specifically, it will create perl.exe,\r |
242 | perl58.dll at the perl toplevel, and various other extension dll's\r |
243 | under the lib\auto directory. If the build fails for any reason, make\r |
244 | sure you have done the previous steps correctly.\r |
245 | \r |
246 | =back\r |
247 | \r |
248 | =head2 Testing Perl on Win32\r |
249 | \r |
250 | Type "dmake test" (or "nmake test"). This will run most of the tests from\r |
251 | the testsuite (many tests will be skipped).\r |
252 | \r |
253 | There should be no test failures when running under Windows NT/2000/XP.\r |
254 | Many tests I<will> fail under Windows 9x due to the inferior command shell.\r |
255 | \r |
256 | Some test failures may occur if you use a command shell other than the\r |
257 | native "cmd.exe", or if you are building from a path that contains\r |
258 | spaces. So don't do that.\r |
259 | \r |
260 | If you are running the tests from a emacs shell window, you may see\r |
261 | failures in op/stat.t. Run "dmake test-notty" in that case.\r |
262 | \r |
263 | If you're using the Borland compiler, you may see a failure in op/taint.t\r |
264 | arising from the inability to find the Borland Runtime DLLs on the system\r |
265 | default path. You will need to copy the DLLs reported by the messages\r |
266 | from where Borland chose to install it, into the Windows system directory\r |
267 | (usually somewhere like C:\WINNT\SYSTEM32) and rerun the test.\r |
268 | \r |
269 | If you're using Borland compiler versions 5.2 and below, you may run into\r |
270 | problems finding the correct header files when building extensions. For\r |
271 | example, building the "Tk" extension may fail because both perl and Tk\r |
272 | contain a header file called "patchlevel.h". The latest Borland compiler\r |
273 | (v5.5) is free of this misbehaviour, and it even supports an\r |
274 | option -VI- for backward (bugward) compatibility for using the old Borland\r |
275 | search algorithm to locate header files.\r |
276 | \r |
277 | Please report any other failures as described under L<BUGS AND CAVEATS>.\r |
278 | \r |
279 | =head2 Installation of Perl on Win32\r |
280 | \r |
281 | Type "dmake install" (or "nmake install"). This will put the newly\r |
282 | built perl and the libraries under whatever C<INST_TOP> points to in the\r |
283 | Makefile. It will also install the pod documentation under\r |
284 | C<$INST_TOP\$VERSION\lib\pod> and HTML versions of the same under\r |
285 | C<$INST_TOP\$VERSION\lib\pod\html>. To use the Perl you just installed,\r |
286 | you will need to add two components to your PATH environment variable,\r |
287 | C<$INST_TOP\$VERSION\bin> and C<$INST_TOP\$VERSION\bin\$ARCHNAME>.\r |
288 | For example:\r |
289 | \r |
290 | set PATH c:\perl\5.6.0\bin;c:\perl\5.6.0\bin\MSWin32-x86;%PATH%\r |
291 | \r |
292 | If you opt to comment out INST_VER and INST_ARCH in the makefiles, the\r |
293 | installation structure is much simpler. In that case, it will be\r |
294 | sufficient to add a single entry to the path, for instance:\r |
295 | \r |
296 | set PATH c:\perl\bin;%PATH%\r |
297 | \r |
298 | =head2 Usage Hints for Perl on Win32\r |
299 | \r |
300 | =over 4\r |
301 | \r |
302 | =item Environment Variables\r |
303 | \r |
304 | The installation paths that you set during the build get compiled\r |
305 | into perl, so you don't have to do anything additional to start\r |
306 | using that perl (except add its location to your PATH variable).\r |
307 | \r |
308 | If you put extensions in unusual places, you can set PERL5LIB\r |
309 | to a list of paths separated by semicolons where you want perl\r |
310 | to look for libraries. Look for descriptions of other environment\r |
311 | variables you can set in L<perlrun>.\r |
312 | \r |
313 | You can also control the shell that perl uses to run system() and\r |
314 | backtick commands via PERL5SHELL. See L<perlrun>.\r |
315 | \r |
316 | Perl does not depend on the registry, but it can look up certain default\r |
317 | values if you choose to put them there. Perl attempts to read entries from\r |
318 | C<HKEY_CURRENT_USER\Software\Perl> and C<HKEY_LOCAL_MACHINE\Software\Perl>.\r |
319 | Entries in the former override entries in the latter. One or more of the\r |
320 | following entries (of type REG_SZ or REG_EXPAND_SZ) may be set:\r |
321 | \r |
322 | lib-$] version-specific standard library path to add to @INC\r |
323 | lib standard library path to add to @INC\r |
324 | sitelib-$] version-specific site library path to add to @INC\r |
325 | sitelib site library path to add to @INC\r |
326 | vendorlib-$] version-specific vendor library path to add to @INC\r |
327 | vendorlib vendor library path to add to @INC\r |
328 | PERL* fallback for all %ENV lookups that begin with "PERL"\r |
329 | \r |
330 | Note the C<$]> in the above is not literal. Substitute whatever version\r |
331 | of perl you want to honor that entry, e.g. C<5.6.0>. Paths must be\r |
332 | separated with semicolons, as usual on win32.\r |
333 | \r |
334 | =item File Globbing\r |
335 | \r |
336 | By default, perl handles file globbing using the File::Glob extension,\r |
337 | which provides portable globbing.\r |
338 | \r |
339 | If you want perl to use globbing that emulates the quirks of DOS\r |
340 | filename conventions, you might want to consider using File::DosGlob\r |
341 | to override the internal glob() implementation. See L<File::DosGlob> for\r |
342 | details.\r |
343 | \r |
344 | =item Using perl from the command line\r |
345 | \r |
346 | If you are accustomed to using perl from various command-line\r |
347 | shells found in UNIX environments, you will be less than pleased\r |
348 | with what Windows offers by way of a command shell.\r |
349 | \r |
350 | The crucial thing to understand about the Windows environment is that\r |
351 | the command line you type in is processed twice before Perl sees it.\r |
352 | First, your command shell (usually CMD.EXE on Windows NT, and\r |
353 | COMMAND.COM on Windows 9x) preprocesses the command line, to handle\r |
354 | redirection, environment variable expansion, and location of the\r |
355 | executable to run. Then, the perl executable splits the remaining\r |
356 | command line into individual arguments, using the C runtime library\r |
357 | upon which Perl was built.\r |
358 | \r |
359 | It is particularly important to note that neither the shell nor the C\r |
360 | runtime do any wildcard expansions of command-line arguments (so\r |
361 | wildcards need not be quoted). Also, the quoting behaviours of the\r |
362 | shell and the C runtime are rudimentary at best (and may, if you are\r |
363 | using a non-standard shell, be inconsistent). The only (useful) quote\r |
364 | character is the double quote ("). It can be used to protect spaces\r |
365 | and other special characters in arguments.\r |
366 | \r |
367 | The Windows NT documentation has almost no description of how the\r |
368 | quoting rules are implemented, but here are some general observations\r |
369 | based on experiments: The C runtime breaks arguments at spaces and\r |
370 | passes them to programs in argc/argv. Double quotes can be used to\r |
371 | prevent arguments with spaces in them from being split up. You can\r |
372 | put a double quote in an argument by escaping it with a backslash and\r |
373 | enclosing the whole argument within double quotes. The backslash and\r |
374 | the pair of double quotes surrounding the argument will be stripped by\r |
375 | the C runtime.\r |
376 | \r |
377 | The file redirection characters "<", ">", and "|" can be quoted by\r |
378 | double quotes (although there are suggestions that this may not always\r |
379 | be true). Single quotes are not treated as quotes by the shell or\r |
380 | the C runtime, they don't get stripped by the shell (just to make\r |
381 | this type of quoting completely useless). The caret "^" has also\r |
382 | been observed to behave as a quoting character, but this appears\r |
383 | to be a shell feature, and the caret is not stripped from the command\r |
384 | line, so Perl still sees it (and the C runtime phase does not treat\r |
385 | the caret as a quote character).\r |
386 | \r |
387 | Here are some examples of usage of the "cmd" shell:\r |
388 | \r |
389 | This prints two doublequotes:\r |
390 | \r |
391 | perl -e "print '\"\"' "\r |
392 | \r |
393 | This does the same:\r |
394 | \r |
395 | perl -e "print \"\\\"\\\"\" "\r |
396 | \r |
397 | This prints "bar" and writes "foo" to the file "blurch":\r |
398 | \r |
399 | perl -e "print 'foo'; print STDERR 'bar'" > blurch\r |
400 | \r |
401 | This prints "foo" ("bar" disappears into nowhereland):\r |
402 | \r |
403 | perl -e "print 'foo'; print STDERR 'bar'" 2> nul\r |
404 | \r |
405 | This prints "bar" and writes "foo" into the file "blurch":\r |
406 | \r |
407 | perl -e "print 'foo'; print STDERR 'bar'" 1> blurch\r |
408 | \r |
409 | This pipes "foo" to the "less" pager and prints "bar" on the console:\r |
410 | \r |
411 | perl -e "print 'foo'; print STDERR 'bar'" | less\r |
412 | \r |
413 | This pipes "foo\nbar\n" to the less pager:\r |
414 | \r |
415 | perl -le "print 'foo'; print STDERR 'bar'" 2>&1 | less\r |
416 | \r |
417 | This pipes "foo" to the pager and writes "bar" in the file "blurch":\r |
418 | \r |
419 | perl -e "print 'foo'; print STDERR 'bar'" 2> blurch | less\r |
420 | \r |
421 | \r |
422 | Discovering the usefulness of the "command.com" shell on Windows 9x\r |
423 | is left as an exercise to the reader :)\r |
424 | \r |
425 | One particularly pernicious problem with the 4NT command shell for\r |
426 | Windows NT is that it (nearly) always treats a % character as indicating\r |
427 | that environment variable expansion is needed. Under this shell, it is\r |
428 | therefore important to always double any % characters which you want\r |
429 | Perl to see (for example, for hash variables), even when they are\r |
430 | quoted.\r |
431 | \r |
432 | =item Building Extensions\r |
433 | \r |
434 | The Comprehensive Perl Archive Network (CPAN) offers a wealth\r |
435 | of extensions, some of which require a C compiler to build.\r |
436 | Look in http://www.cpan.org/ for more information on CPAN.\r |
437 | \r |
438 | Note that not all of the extensions available from CPAN may work\r |
439 | in the Win32 environment; you should check the information at\r |
440 | http://testers.cpan.org/ before investing too much effort into\r |
441 | porting modules that don't readily build.\r |
442 | \r |
443 | Most extensions (whether they require a C compiler or not) can\r |
444 | be built, tested and installed with the standard mantra:\r |
445 | \r |
446 | perl Makefile.PL\r |
447 | $MAKE\r |
448 | $MAKE test\r |
449 | $MAKE install\r |
450 | \r |
451 | where $MAKE is whatever 'make' program you have configured perl to\r |
452 | use. Use "perl -V:make" to find out what this is. Some extensions\r |
453 | may not provide a testsuite (so "$MAKE test" may not do anything or\r |
454 | fail), but most serious ones do.\r |
455 | \r |
456 | It is important that you use a supported 'make' program, and\r |
457 | ensure Config.pm knows about it. If you don't have nmake, you can\r |
458 | either get dmake from the location mentioned earlier or get an\r |
459 | old version of nmake reportedly available from:\r |
460 | \r |
461 | ftp://ftp.microsoft.com/Softlib/MSLFILES/nmake15.exe\r |
462 | \r |
463 | Another option is to use the make written in Perl, available from\r |
464 | CPAN.\r |
465 | \r |
466 | http://www.cpan.org/modules/by-module/Make/\r |
467 | \r |
468 | You may also use dmake. See L</"Make"> above on how to get it.\r |
469 | \r |
470 | Note that MakeMaker actually emits makefiles with different syntax\r |
471 | depending on what 'make' it thinks you are using. Therefore, it is\r |
472 | important that one of the following values appears in Config.pm:\r |
473 | \r |
474 | make='nmake' # MakeMaker emits nmake syntax\r |
475 | make='dmake' # MakeMaker emits dmake syntax\r |
476 | any other value # MakeMaker emits generic make syntax\r |
477 | (e.g GNU make, or Perl make)\r |
478 | \r |
479 | If the value doesn't match the 'make' program you want to use,\r |
480 | edit Config.pm to fix it.\r |
481 | \r |
482 | If a module implements XSUBs, you will need one of the supported\r |
483 | C compilers. You must make sure you have set up the environment for\r |
484 | the compiler for command-line compilation.\r |
485 | \r |
486 | If a module does not build for some reason, look carefully for\r |
487 | why it failed, and report problems to the module author. If\r |
488 | it looks like the extension building support is at fault, report\r |
489 | that with full details of how the build failed using the perlbug\r |
490 | utility.\r |
491 | \r |
492 | =item Command-line Wildcard Expansion\r |
493 | \r |
494 | The default command shells on DOS descendant operating systems (such\r |
495 | as they are) usually do not expand wildcard arguments supplied to\r |
496 | programs. They consider it the application's job to handle that.\r |
497 | This is commonly achieved by linking the application (in our case,\r |
498 | perl) with startup code that the C runtime libraries usually provide.\r |
499 | However, doing that results in incompatible perl versions (since the\r |
500 | behavior of the argv expansion code differs depending on the\r |
501 | compiler, and it is even buggy on some compilers). Besides, it may\r |
502 | be a source of frustration if you use such a perl binary with an\r |
503 | alternate shell that *does* expand wildcards.\r |
504 | \r |
505 | Instead, the following solution works rather well. The nice things\r |
506 | about it are 1) you can start using it right away; 2) it is more \r |
507 | powerful, because it will do the right thing with a pattern like\r |
508 | */*/*.c; 3) you can decide whether you do/don't want to use it; and\r |
509 | 4) you can extend the method to add any customizations (or even \r |
510 | entirely different kinds of wildcard expansion).\r |
511 | \r |
512 | C:\> copy con c:\perl\lib\Wild.pm\r |
513 | # Wild.pm - emulate shell @ARGV expansion on shells that don't\r |
514 | use File::DosGlob;\r |
515 | @ARGV = map {\r |
516 | my @g = File::DosGlob::glob($_) if /[*?]/;\r |
517 | @g ? @g : $_;\r |
518 | } @ARGV;\r |
519 | 1;\r |
520 | ^Z\r |
521 | C:\> set PERL5OPT=-MWild\r |
522 | C:\> perl -le "for (@ARGV) { print }" */*/perl*.c\r |
523 | p4view/perl/perl.c\r |
524 | p4view/perl/perlio.c\r |
525 | p4view/perl/perly.c\r |
526 | perl5.005/win32/perlglob.c\r |
527 | perl5.005/win32/perllib.c\r |
528 | perl5.005/win32/perlglob.c\r |
529 | perl5.005/win32/perllib.c\r |
530 | perl5.005/win32/perlglob.c\r |
531 | perl5.005/win32/perllib.c\r |
532 | \r |
533 | Note there are two distinct steps there: 1) You'll have to create\r |
534 | Wild.pm and put it in your perl lib directory. 2) You'll need to\r |
535 | set the PERL5OPT environment variable. If you want argv expansion\r |
536 | to be the default, just set PERL5OPT in your default startup\r |
537 | environment.\r |
538 | \r |
539 | If you are using the Visual C compiler, you can get the C runtime's\r |
540 | command line wildcard expansion built into perl binary. The resulting\r |
541 | binary will always expand unquoted command lines, which may not be\r |
542 | what you want if you use a shell that does that for you. The expansion\r |
543 | done is also somewhat less powerful than the approach suggested above.\r |
544 | \r |
545 | =item Win32 Specific Extensions\r |
546 | \r |
547 | A number of extensions specific to the Win32 platform are available\r |
548 | from CPAN. You may find that many of these extensions are meant to\r |
549 | be used under the Activeware port of Perl, which used to be the only\r |
550 | native port for the Win32 platform. Since the Activeware port does not\r |
551 | have adequate support for Perl's extension building tools, these\r |
552 | extensions typically do not support those tools either and, therefore,\r |
553 | cannot be built using the generic steps shown in the previous section.\r |
554 | \r |
555 | To ensure smooth transitioning of existing code that uses the\r |
556 | ActiveState port, there is a bundle of Win32 extensions that contains\r |
557 | all of the ActiveState extensions and most other Win32 extensions from\r |
558 | CPAN in source form, along with many added bugfixes, and with MakeMaker\r |
559 | support. This bundle is available at:\r |
560 | \r |
561 | http://www.cpan.org/authors/id/GSAR/libwin32-0.18.zip\r |
562 | \r |
563 | See the README in that distribution for building and installation\r |
564 | instructions. Look for later versions that may be available at the\r |
565 | same location.\r |
566 | \r |
567 | =item Notes on 64-bit Windows\r |
568 | \r |
569 | Windows .NET Server supports the LLP64 data model on the Intel Itanium\r |
570 | architecture.\r |
571 | \r |
572 | The LLP64 data model is different from the LP64 data model that is the\r |
573 | norm on 64-bit Unix platforms. In the former, C<int> and C<long> are\r |
574 | both 32-bit data types, while pointers are 64 bits wide. In addition,\r |
575 | there is a separate 64-bit wide integral type, C<__int64>. In contrast,\r |
576 | the LP64 data model that is pervasive on Unix platforms provides C<int>\r |
577 | as the 32-bit type, while both the C<long> type and pointers are of\r |
578 | 64-bit precision. Note that both models provide for 64-bits of\r |
579 | addressability.\r |
580 | \r |
581 | 64-bit Windows running on Itanium is capable of running 32-bit x86\r |
582 | binaries transparently. This means that you could use a 32-bit build\r |
583 | of Perl on a 64-bit system. Given this, why would one want to build\r |
584 | a 64-bit build of Perl? Here are some reasons why you would bother:\r |
585 | \r |
586 | =item *\r |
587 | \r |
588 | A 64-bit native application will run much more efficiently on\r |
589 | Itanium hardware.\r |
590 | \r |
591 | =item *\r |
592 | \r |
593 | There is no 2GB limit on process size.\r |
594 | \r |
595 | =item *\r |
596 | \r |
597 | Perl automatically provides large file support when built under\r |
598 | 64-bit Windows.\r |
599 | \r |
600 | =item *\r |
601 | \r |
602 | Embedding Perl inside a 64-bit application.\r |
603 | \r |
604 | =back\r |
605 | \r |
606 | =head2 Running Perl Scripts\r |
607 | \r |
608 | Perl scripts on UNIX use the "#!" (a.k.a "shebang") line to\r |
609 | indicate to the OS that it should execute the file using perl.\r |
610 | Win32 has no comparable means to indicate arbitrary files are\r |
611 | executables.\r |
612 | \r |
613 | Instead, all available methods to execute plain text files on\r |
614 | Win32 rely on the file "extension". There are three methods\r |
615 | to use this to execute perl scripts:\r |
616 | \r |
617 | =over 8\r |
618 | \r |
619 | =item 1\r |
620 | \r |
621 | There is a facility called "file extension associations" that will\r |
622 | work in Windows NT 4.0. This can be manipulated via the two\r |
623 | commands "assoc" and "ftype" that come standard with Windows NT\r |
624 | 4.0. Type "ftype /?" for a complete example of how to set this\r |
625 | up for perl scripts (Say what? You thought Windows NT wasn't\r |
626 | perl-ready? :).\r |
627 | \r |
628 | =item 2\r |
629 | \r |
630 | Since file associations don't work everywhere, and there are\r |
631 | reportedly bugs with file associations where it does work, the\r |
632 | old method of wrapping the perl script to make it look like a\r |
633 | regular batch file to the OS, may be used. The install process\r |
634 | makes available the "pl2bat.bat" script which can be used to wrap\r |
635 | perl scripts into batch files. For example:\r |
636 | \r |
637 | pl2bat foo.pl\r |
638 | \r |
639 | will create the file "FOO.BAT". Note "pl2bat" strips any\r |
640 | .pl suffix and adds a .bat suffix to the generated file.\r |
641 | \r |
642 | If you use the 4DOS/NT or similar command shell, note that\r |
643 | "pl2bat" uses the "%*" variable in the generated batch file to\r |
644 | refer to all the command line arguments, so you may need to make\r |
645 | sure that construct works in batch files. As of this writing,\r |
646 | 4DOS/NT users will need a "ParameterChar = *" statement in their\r |
647 | 4NT.INI file or will need to execute "setdos /p*" in the 4DOS/NT\r |
648 | startup file to enable this to work.\r |
649 | \r |
650 | =item 3\r |
651 | \r |
652 | Using "pl2bat" has a few problems: the file name gets changed,\r |
653 | so scripts that rely on C<$0> to find what they must do may not\r |
654 | run properly; running "pl2bat" replicates the contents of the\r |
655 | original script, and so this process can be maintenance intensive\r |
656 | if the originals get updated often. A different approach that\r |
657 | avoids both problems is possible.\r |
658 | \r |
659 | A script called "runperl.bat" is available that can be copied\r |
660 | to any filename (along with the .bat suffix). For example,\r |
661 | if you call it "foo.bat", it will run the file "foo" when it is\r |
662 | executed. Since you can run batch files on Win32 platforms simply\r |
663 | by typing the name (without the extension), this effectively\r |
664 | runs the file "foo", when you type either "foo" or "foo.bat".\r |
665 | With this method, "foo.bat" can even be in a different location\r |
666 | than the file "foo", as long as "foo" is available somewhere on\r |
667 | the PATH. If your scripts are on a filesystem that allows symbolic\r |
668 | links, you can even avoid copying "runperl.bat".\r |
669 | \r |
670 | Here's a diversion: copy "runperl.bat" to "runperl", and type\r |
671 | "runperl". Explain the observed behavior, or lack thereof. :)\r |
672 | Hint: .gnidnats llits er'uoy fi ,"lrepnur" eteled :tniH\r |
673 | \r |
674 | =item Miscellaneous Things\r |
675 | \r |
676 | A full set of HTML documentation is installed, so you should be\r |
677 | able to use it if you have a web browser installed on your\r |
678 | system.\r |
679 | \r |
680 | C<perldoc> is also a useful tool for browsing information contained\r |
681 | in the documentation, especially in conjunction with a pager\r |
682 | like C<less> (recent versions of which have Win32 support). You may\r |
683 | have to set the PAGER environment variable to use a specific pager.\r |
684 | "perldoc -f foo" will print information about the perl operator\r |
685 | "foo".\r |
686 | \r |
687 | If you find bugs in perl, you can run C<perlbug> to create a\r |
688 | bug report (you may have to send it manually if C<perlbug> cannot\r |
689 | find a mailer on your system).\r |
690 | \r |
691 | =back\r |
692 | \r |
693 | =head1 BUGS AND CAVEATS\r |
694 | \r |
695 | Norton AntiVirus interferes with the build process, particularly if \r |
696 | set to "AutoProtect, All Files, when Opened". Unlike large applications \r |
697 | the perl build process opens and modifies a lot of files. Having the \r |
698 | the AntiVirus scan each and every one slows build the process significantly.\r |
699 | Worse, with PERLIO=stdio the build process fails with peculiar messages\r |
700 | as the virus checker interacts badly with miniperl.exe writing configure \r |
701 | files (it seems to either catch file part written and treat it as suspicious,\r |
702 | or virus checker may have it "locked" in a way which inhibits miniperl\r |
703 | updating it). The build does complete with \r |
704 | \r |
705 | set PERLIO=perlio\r |
706 | \r |
707 | but that may be just luck. Other AntiVirus software may have similar issues.\r |
708 | \r |
709 | Some of the built-in functions do not act exactly as documented in\r |
710 | L<perlfunc>, and a few are not implemented at all. To avoid\r |
711 | surprises, particularly if you have had prior exposure to Perl\r |
712 | in other operating environments or if you intend to write code\r |
713 | that will be portable to other environments. See L<perlport>\r |
714 | for a reasonably definitive list of these differences.\r |
715 | \r |
716 | Not all extensions available from CPAN may build or work properly\r |
717 | in the Win32 environment. See L</"Building Extensions">.\r |
718 | \r |
719 | Most C<socket()> related calls are supported, but they may not\r |
720 | behave as on Unix platforms. See L<perlport> for the full list.\r |
721 | \r |
722 | Signal handling may not behave as on Unix platforms (where it\r |
723 | doesn't exactly "behave", either :). For instance, calling C<die()>\r |
724 | or C<exit()> from signal handlers will cause an exception, since most\r |
725 | implementations of C<signal()> on Win32 are severely crippled.\r |
726 | Thus, signals may work only for simple things like setting a flag\r |
727 | variable in the handler. Using signals under this port should\r |
728 | currently be considered unsupported.\r |
729 | \r |
730 | Please send detailed descriptions of any problems and solutions that \r |
731 | you may find to <F<perlbug@perl.com>>, along with the output produced\r |
732 | by C<perl -V>.\r |
733 | \r |
734 | =head1 AUTHORS\r |
735 | \r |
736 | =over 4\r |
737 | \r |
738 | =item Gary Ng E<lt>71564.1743@CompuServe.COME<gt>\r |
739 | \r |
740 | =item Gurusamy Sarathy E<lt>gsar@activestate.comE<gt>\r |
741 | \r |
742 | =item Nick Ing-Simmons E<lt>nick@ing-simmons.netE<gt>\r |
743 | \r |
744 | =back\r |
745 | \r |
746 | This document is maintained by Gurusamy Sarathy.\r |
747 | \r |
748 | =head1 SEE ALSO\r |
749 | \r |
750 | L<perl>\r |
751 | \r |
752 | =head1 HISTORY\r |
753 | \r |
754 | This port was originally contributed by Gary Ng around 5.003_24,\r |
755 | and borrowed from the Hip Communications port that was available\r |
756 | at the time. Various people have made numerous and sundry hacks\r |
757 | since then.\r |
758 | \r |
759 | Borland support was added in 5.004_01 (Gurusamy Sarathy).\r |
760 | \r |
761 | GCC/mingw32 support was added in 5.005 (Nick Ing-Simmons).\r |
762 | \r |
763 | Support for PERL_OBJECT was added in 5.005 (ActiveState Tool Corp).\r |
764 | \r |
765 | Support for fork() emulation was added in 5.6 (ActiveState Tool Corp).\r |
766 | \r |
767 | Win9x support was added in 5.6 (Benjamin Stuhl).\r |
768 | \r |
769 | Support for 64-bit Windows added in 5.8 (ActiveState Corp).\r |
770 | \r |
771 | Last updated: 20 April 2002\r |
772 | \r |
773 | =cut\r |