3 perlport - Writing portable Perl
7 Perl runs on numerous operating systems. While most of them share
8 much in common, they also have their own unique features.
10 This document is meant to help you to find out what constitutes portable
11 Perl code. That way once you make a decision to write portably,
12 you know where the lines are drawn, and you can stay within them.
14 There is a tradeoff between taking full advantage of one particular
15 type of computer and taking advantage of a full range of them.
16 Naturally, as you broaden your range and become more diverse, the
17 common factors drop, and you are left with an increasingly smaller
18 area of common ground in which you can operate to accomplish a
19 particular task. Thus, when you begin attacking a problem, it is
20 important to consider under which part of the tradeoff curve you
21 want to operate. Specifically, you must decide whether it is
22 important that the task that you are coding have the full generality
23 of being portable, or whether to just get the job done right now.
24 This is the hardest choice to be made. The rest is easy, because
25 Perl provides many choices, whichever way you want to approach your
28 Looking at it another way, writing portable code is usually about
29 willfully limiting your available choices. Naturally, it takes
30 discipline and sacrifice to do that. The product of portability
31 and convenience may be a constant. You have been warned.
33 Be aware of two important points:
37 =item Not all Perl programs have to be portable
39 There is no reason you should not use Perl as a language to glue Unix
40 tools together, or to prototype a Macintosh application, or to manage the
41 Windows registry. If it makes no sense to aim for portability for one
42 reason or another in a given program, then don't bother.
44 =item Nearly all of Perl already I<is> portable
46 Don't be fooled into thinking that it is hard to create portable Perl
47 code. It isn't. Perl tries its level-best to bridge the gaps between
48 what's available on different platforms, and all the means available to
49 use those features. Thus almost all Perl code runs on any machine
50 without modification. But there are some significant issues in
51 writing portable code, and this document is entirely about those issues.
55 Here's the general rule: When you approach a task commonly done
56 using a whole range of platforms, think about writing portable
57 code. That way, you don't sacrifice much by way of the implementation
58 choices you can avail yourself of, and at the same time you can give
59 your users lots of platform choices. On the other hand, when you have to
60 take advantage of some unique feature of a particular platform, as is
61 often the case with systems programming (whether for Unix, Windows,
62 S<Mac OS>, VMS, etc.), consider writing platform-specific code.
64 When the code will run on only two or three operating systems, you
65 may need to consider only the differences of those particular systems.
66 The important thing is to decide where the code will run and to be
67 deliberate in your decision.
69 The material below is separated into three main sections: main issues of
70 portability (L<"ISSUES">, platform-specific issues (L<"PLATFORMS">, and
71 built-in perl functions that behave differently on various ports
72 (L<"FUNCTION IMPLEMENTATIONS">.
74 This information should not be considered complete; it includes possibly
75 transient information about idiosyncrasies of some of the ports, almost
76 all of which are in a state of constant evolution. Thus, this material
77 should be considered a perpetual work in progress
78 (<IMG SRC="yellow_sign.gif" ALT="Under Construction">).
84 In most operating systems, lines in files are terminated by newlines.
85 Just what is used as a newline may vary from OS to OS. Unix
86 traditionally uses C<\012>, one type of DOSish I/O uses C<\015\012>,
87 and S<Mac OS> uses C<\015>.
89 Perl uses C<\n> to represent the "logical" newline, where what is
90 logical may depend on the platform in use. In MacPerl, C<\n> always
91 means C<\015>. In DOSish perls, C<\n> usually means C<\012>, but
92 when accessing a file in "text" mode, STDIO translates it to (or
93 from) C<\015\012>, depending on whether you're reading or writing.
94 Unix does the same thing on ttys in canonical mode. C<\015\012>
95 is commonly referred to as CRLF.
97 Because of the "text" mode translation, DOSish perls have limitations
98 in using C<seek> and C<tell> on a file accessed in "text" mode.
99 Stick to C<seek>-ing to locations you got from C<tell> (and no
100 others), and you are usually free to use C<seek> and C<tell> even
101 in "text" mode. Using C<seek> or C<tell> or other file operations
102 may be non-portable. If you use C<binmode> on a file, however, you
103 can usually C<seek> and C<tell> with arbitrary values in safety.
105 A common misconception in socket programming is that C<\n> eq C<\012>
106 everywhere. When using protocols such as common Internet protocols,
107 C<\012> and C<\015> are called for specifically, and the values of
108 the logical C<\n> and C<\r> (carriage return) are not reliable.
110 print SOCKET "Hi there, client!\r\n"; # WRONG
111 print SOCKET "Hi there, client!\015\012"; # RIGHT
113 However, using C<\015\012> (or C<\cM\cJ>, or C<\x0D\x0A>) can be tedious
114 and unsightly, as well as confusing to those maintaining the code. As
115 such, the Socket module supplies the Right Thing for those who want it.
117 use Socket qw(:DEFAULT :crlf);
118 print SOCKET "Hi there, client!$CRLF" # RIGHT
120 When reading from a socket, remember that the default input record
121 separator C<$/> is C<\n>, but robust socket code will recognize as
122 either C<\012> or C<\015\012> as end of line:
128 Because both CRLF and LF end in LF, the input record separator can
129 be set to LF and any CR stripped later. Better to write:
131 use Socket qw(:DEFAULT :crlf);
132 local($/) = LF; # not needed if $/ is already \012
135 s/$CR?$LF/\n/; # not sure if socket uses LF or CRLF, OK
136 # s/\015?\012/\n/; # same thing
139 This example is preferred over the previous one--even for Unix
140 platforms--because now any C<\015>'s (C<\cM>'s) are stripped out
141 (and there was much rejoicing).
143 Similarly, functions that return text data--such as a function that
144 fetches a web page--should sometimes translate newlines before
145 returning the data, if they've not yet been translated to the local
146 newline representation. A single line of code will often suffice:
148 $data =~ s/\015?\012/\n/g;
151 Some of this may be confusing. Here's a handy reference to the ASCII CR
152 and LF characters. You can print it out and stick it in your wallet.
154 LF == \012 == \x0A == \cJ == ASCII 10
155 CR == \015 == \x0D == \cM == ASCII 13
158 ---------------------------
161 \n * | LF | CRLF | CR |
162 \r * | CR | CR | LF |
163 ---------------------------
166 The Unix column assumes that you are not accessing a serial line
167 (like a tty) in canonical mode. If you are, then CR on input becomes
168 "\n", and "\n" on output becomes CRLF.
170 These are just the most common definitions of C<\n> and C<\r> in Perl.
171 There may well be others.
173 =head2 Numbers endianness and Width
175 Different CPUs store integers and floating point numbers in different
176 orders (called I<endianness>) and widths (32-bit and 64-bit being the
177 most common today). This affects your programs when they attempt to transfer
178 numbers in binary format from one CPU architecture to another,
179 usually either "live" via network connection, or by storing the
180 numbers to secondary storage such as a disk file or tape.
182 Conflicting storage orders make utter mess out of the numbers. If a
183 little-endian host (Intel, VAX) stores 0x12345678 (305419896 in
184 decimal), a big-endian host (Motorola, Sparc, PA) reads it as
185 0x78563412 (2018915346 in decimal). Alpha and MIPS can be either:
186 Digital/Compaq used/uses them in little-endian mode; SGI/Cray uses
187 them in big-endian mode. To avoid this problem in network (socket)
188 connections use the C<pack> and C<unpack> formats C<n> and C<N>, the
189 "network" orders. These are guaranteed to be portable.
191 You can explore the endianness of your platform by unpacking a
192 data structure packed in native format such as:
194 print unpack("h*", pack("s2", 1, 2)), "\n";
195 # '10002000' on e.g. Intel x86 or Alpha 21064 in little-endian mode
196 # '00100020' on e.g. Motorola 68040
198 If you need to distinguish between endian architectures you could use
199 either of the variables set like so:
201 $is_big_endian = unpack("h*", pack("s", 1)) =~ /01/;
202 $is_little_endian = unpack("h*", pack("s", 1)) =~ /^1/;
204 Differing widths can cause truncation even between platforms of equal
205 endianness. The platform of shorter width loses the upper parts of the
206 number. There is no good solution for this problem except to avoid
207 transferring or storing raw binary numbers.
209 One can circumnavigate both these problems in two ways. Either
210 transfer and store numbers always in text format, instead of raw
211 binary, or else consider using modules like Data::Dumper (included in
212 the standard distribution as of Perl 5.005) and Storable. Keeping
213 all data as text significantly simplifies matters.
215 =head2 Files and Filesystems
217 Most platforms these days structure files in a hierarchical fashion.
218 So, it is reasonably safe to assume that all platforms support the
219 notion of a "path" to uniquely identify a file on the system. How
220 that path is really written, though, differs considerably.
222 Although similar, file path specifications differ between Unix,
223 Windows, S<Mac OS>, OS/2, VMS, VOS, S<RISC OS>, and probably others.
224 Unix, for example, is one of the few OSes that has the elegant idea
225 of a single root directory.
227 DOS, OS/2, VMS, VOS, and Windows can work similarly to Unix with C</>
228 as path separator, or in their own idiosyncratic ways (such as having
229 several root directories and various "unrooted" device files such NIL:
232 S<Mac OS> uses C<:> as a path separator instead of C</>.
234 The filesystem may support neither hard links (C<link>) nor
235 symbolic links (C<symlink>, C<readlink>, C<lstat>).
237 The filesystem may support neither access timestamp nor change
238 timestamp (meaning that about the only portable timestamp is the
239 modification timestamp), or one second granularity of any timestamps
240 (e.g. the FAT filesystem limits the time granularity to two seconds).
242 VOS perl can emulate Unix filenames with C</> as path separator. The
243 native pathname characters greater-than, less-than, number-sign, and
244 percent-sign are always accepted.
246 S<RISC OS> perl can emulate Unix filenames with C</> as path
247 separator, or go native and use C<.> for path separator and C<:> to
248 signal filesystems and disk names.
250 If all this is intimidating, have no (well, maybe only a little)
251 fear. There are modules that can help. The File::Spec modules
252 provide methods to do the Right Thing on whatever platform happens
253 to be running the program.
255 use File::Spec::Functions;
256 chdir(updir()); # go up one directory
257 $file = catfile(curdir(), 'temp', 'file.txt');
258 # on Unix and Win32, './temp/file.txt'
259 # on Mac OS, ':temp:file.txt'
260 # on VMS, '[.temp]file.txt'
262 File::Spec is available in the standard distribution as of version
263 5.004_05. File::Spec::Functions is only in File::Spec 0.7 and later,
264 and some versions of perl come with version 0.6. If File::Spec
265 is not updated to 0.7 or later, you must use the object-oriented
266 interface from File::Spec (or upgrade File::Spec).
268 In general, production code should not have file paths hardcoded.
269 Making them user-supplied or read from a configuration file is
270 better, keeping in mind that file path syntax varies on different
273 This is especially noticeable in scripts like Makefiles and test suites,
274 which often assume C</> as a path separator for subdirectories.
276 Also of use is File::Basename from the standard distribution, which
277 splits a pathname into pieces (base filename, full path to directory,
280 Even when on a single platform (if you can call Unix a single platform),
281 remember not to count on the existence or the contents of particular
282 system-specific files or directories, like F</etc/passwd>,
283 F</etc/sendmail.conf>, F</etc/resolv.conf>, or even F</tmp/>. For
284 example, F</etc/passwd> may exist but not contain the encrypted
285 passwords, because the system is using some form of enhanced security.
286 Or it may not contain all the accounts, because the system is using NIS.
287 If code does need to rely on such a file, include a description of the
288 file and its format in the code's documentation, then make it easy for
289 the user to override the default location of the file.
291 Don't assume a text file will end with a newline. They should,
294 Do not have two files of the same name with different case, like
295 F<test.pl> and F<Test.pl>, as many platforms have case-insensitive
296 filenames. Also, try not to have non-word characters (except for C<.>)
297 in the names, and keep them to the 8.3 convention, for maximum
298 portability, onerous a burden though this may appear.
300 Likewise, when using the AutoSplit module, try to keep your functions to
301 8.3 naming and case-insensitive conventions; or, at the least,
302 make it so the resulting files have a unique (case-insensitively)
305 Whitespace in filenames is tolerated on most systems, but not all.
306 Many systems (DOS, VMS) cannot have more than one C<.> in their filenames.
308 Don't assume C<< > >> won't be the first character of a filename.
309 Always use C<< < >> explicitly to open a file for reading,
310 unless you want the user to be able to specify a pipe open.
312 open(FILE, "< $existing_file") or die $!;
314 If filenames might use strange characters, it is safest to open it
315 with C<sysopen> instead of C<open>. C<open> is magic and can
316 translate characters like C<< > >>, C<< < >>, and C<|>, which may
317 be the wrong thing to do. (Sometimes, though, it's the right thing.)
319 =head2 System Interaction
321 Not all platforms provide a command line. These are usually platforms
322 that rely primarily on a Graphical User Interface (GUI) for user
323 interaction. A program requiring a command line interface might
324 not work everywhere. This is probably for the user of the program
325 to deal with, so don't stay up late worrying about it.
327 Some platforms can't delete or rename files held open by the system.
328 Remember to C<close> files when you are done with them. Don't
329 C<unlink> or C<rename> an open file. Don't C<tie> or C<open> a
330 file already tied or opened; C<untie> or C<close> it first.
332 Don't open the same file more than once at a time for writing, as some
333 operating systems put mandatory locks on such files.
335 Don't count on a specific environment variable existing in C<%ENV>.
336 Don't count on C<%ENV> entries being case-sensitive, or even
339 Don't count on signals or C<%SIG> for anything.
341 Don't count on filename globbing. Use C<opendir>, C<readdir>, and
344 Don't count on per-program environment variables, or per-program current
347 Don't count on specific values of C<$!>.
349 =head2 Interprocess Communication (IPC)
351 In general, don't directly access the system in code meant to be
352 portable. That means, no C<system>, C<exec>, C<fork>, C<pipe>,
353 C<``>, C<qx//>, C<open> with a C<|>, nor any of the other things
354 that makes being a perl hacker worth being.
356 Commands that launch external processes are generally supported on
357 most platforms (though many of them do not support any type of
358 forking). The problem with using them arises from what you invoke
359 them on. External tools are often named differently on different
360 platforms, may not be available in the same location, might accept
361 different arguments, can behave differently, and often present their
362 results in a platform-dependent way. Thus, you should seldom depend
363 on them to produce consistent results. (Then again, if you're calling
364 I<netstat -a>, you probably don't expect it to run on both Unix and CP/M.)
366 One especially common bit of Perl code is opening a pipe to B<sendmail>:
368 open(MAIL, '|/usr/lib/sendmail -t')
369 or die "cannot fork sendmail: $!";
371 This is fine for systems programming when sendmail is known to be
372 available. But it is not fine for many non-Unix systems, and even
373 some Unix systems that may not have sendmail installed. If a portable
374 solution is needed, see the various distributions on CPAN that deal
375 with it. Mail::Mailer and Mail::Send in the MailTools distribution are
376 commonly used, and provide several mailing methods, including mail,
377 sendmail, and direct SMTP (via Net::SMTP) if a mail transfer agent is
378 not available. Mail::Sendmail is a standalone module that provides
379 simple, platform-independent mailing.
381 The Unix System V IPC (C<msg*(), sem*(), shm*()>) is not available
382 even on all Unix platforms.
384 The rule of thumb for portable code is: Do it all in portable Perl, or
385 use a module (that may internally implement it with platform-specific
386 code, but expose a common interface).
388 =head2 External Subroutines (XS)
390 XS code can usually be made to work with any platform, but dependent
391 libraries, header files, etc., might not be readily available or
392 portable, or the XS code itself might be platform-specific, just as Perl
393 code might be. If the libraries and headers are portable, then it is
394 normally reasonable to make sure the XS code is portable, too.
396 A different type of portability issue arises when writing XS code:
397 availability of a C compiler on the end-user's system. C brings
398 with it its own portability issues, and writing XS code will expose
399 you to some of those. Writing purely in Perl is an easier way to
402 =head2 Standard Modules
404 In general, the standard modules work across platforms. Notable
405 exceptions are the CPAN module (which currently makes connections to external
406 programs that may not be available), platform-specific modules (like
407 ExtUtils::MM_VMS), and DBM modules.
409 There is no one DBM module available on all platforms.
410 SDBM_File and the others are generally available on all Unix and DOSish
411 ports, but not in MacPerl, where only NBDM_File and DB_File are
414 The good news is that at least some DBM module should be available, and
415 AnyDBM_File will use whichever module it can find. Of course, then
416 the code needs to be fairly strict, dropping to the greatest common
417 factor (e.g., not exceeding 1K for each record), so that it will
418 work with any DBM module. See L<AnyDBM_File> for more details.
422 The system's notion of time of day and calendar date is controlled in
423 widely different ways. Don't assume the timezone is stored in C<$ENV{TZ}>,
424 and even if it is, don't assume that you can control the timezone through
427 Don't assume that the epoch starts at 00:00:00, January 1, 1970,
428 because that is OS- and implementation-specific. It is better to store a date
429 in an unambiguous representation. The ISO-8601 standard defines
430 "YYYY-MM-DD" as the date format. A text representation (like "1987-12-18")
431 can be easily converted into an OS-specific value using a module like
432 Date::Parse. An array of values, such as those returned by
433 C<localtime>, can be converted to an OS-specific representation using
436 When calculating specific times, such as for tests in time or date modules,
437 it may be appropriate to calculate an offset for the epoch.
440 $offset = Time::Local::timegm(0, 0, 0, 1, 0, 70);
442 The value for C<$offset> in Unix will be C<0>, but in Mac OS will be
443 some large number. C<$offset> can then be added to a Unix time value
444 to get what should be the proper value on any system.
446 =head2 Character sets and character encoding
448 Assume little about character sets. Assume nothing about
449 numerical values (C<ord>, C<chr>) of characters. Do not
450 assume that the alphabetic characters are encoded contiguously (in
451 the numeric sense). Do not assume anything about the ordering of the
452 characters. The lowercase letters may come before or after the
453 uppercase letters; the lowercase and uppercase may be interlaced so
454 that both `a' and `A' come before `b'; the accented and other
455 international characters may be interlaced so that E<auml> comes
458 =head2 Internationalisation
460 If you may assume POSIX (a rather large assumption), you may read
461 more about the POSIX locale system from L<perllocale>. The locale
462 system at least attempts to make things a little bit more portable,
463 or at least more convenient and native-friendly for non-English
464 users. The system affects character sets and encoding, and date
465 and time formatting--amongst other things.
467 =head2 System Resources
469 If your code is destined for systems with severely constrained (or
470 missing!) virtual memory systems then you want to be I<especially> mindful
471 of avoiding wasteful constructs such as:
473 # NOTE: this is no longer "bad" in perl5.005
474 for (0..10000000) {} # bad
475 for (my $x = 0; $x <= 10000000; ++$x) {} # good
477 @lines = <VERY_LARGE_FILE>; # bad
479 while (<FILE>) {$file .= $_} # sometimes bad
480 $file = join('', <FILE>); # better
482 The last two constructs may appear unintuitive to most people. The
483 first repeatedly grows a string, whereas the second allocates a
484 large chunk of memory in one go. On some systems, the second is
485 more efficient that the first.
489 Most multi-user platforms provide basic levels of security, usually
490 implemented at the filesystem level. Some, however, do
491 not--unfortunately. Thus the notion of user id, or "home" directory,
492 or even the state of being logged-in, may be unrecognizable on many
493 platforms. If you write programs that are security-conscious, it
494 is usually best to know what type of system you will be running
495 under so that you can write code explicitly for that platform (or
500 For those times when it is necessary to have platform-specific code,
501 consider keeping the platform-specific code in one place, making porting
502 to other platforms easier. Use the Config module and the special
503 variable C<$^O> to differentiate platforms, as described in
506 Be careful in the tests you supply with your module or programs.
507 Module code may be fully portable, but its tests might not be. This
508 often happens when tests spawn off other processes or call external
509 programs to aid in the testing, or when (as noted above) the tests
510 assume certain things about the filesystem and paths. Be careful
511 not to depend on a specific output style for errors, such as when
512 checking C<$!> after an system call. Some platforms expect a certain
513 output format, and perl on those platforms may have been adjusted
514 accordingly. Most specifically, don't anchor a regex when testing
519 Modules uploaded to CPAN are tested by a variety of volunteers on
520 different platforms. These CPAN testers are notified by mail of each
521 new upload, and reply to the list with PASS, FAIL, NA (not applicable to
522 this platform), or UNKNOWN (unknown), along with any relevant notations.
524 The purpose of the testing is twofold: one, to help developers fix any
525 problems in their code that crop up because of lack of testing on other
526 platforms; two, to provide users with information about whether
527 a given module works on a given platform.
531 =item Mailing list: cpan-testers@perl.org
533 =item Testing results: http://testers.cpan.org/
539 As of version 5.002, Perl is built with a C<$^O> variable that
540 indicates the operating system it was built on. This was implemented
541 to help speed up code that would otherwise have to C<use Config>
542 and use the value of C<$Config{osname}>. Of course, to get more
543 detailed information about the system, looking into C<%Config> is
544 certainly recommended.
546 C<%Config> cannot always be trusted, however, because it was built
547 at compile time. If perl was built in one place, then transferred
548 elsewhere, some values may be wrong. The values may even have been
549 edited after the fact.
553 Perl works on a bewildering variety of Unix and Unix-like platforms (see
554 e.g. most of the files in the F<hints/> directory in the source code kit).
555 On most of these systems, the value of C<$^O> (hence C<$Config{'osname'}>,
556 too) is determined either by lowercasing and stripping punctuation from the
557 first field of the string returned by typing C<uname -a> (or a similar command)
558 at the shell prompt or by testing the file system for the presence of
559 uniquely named files such as a kernel or header file. Here, for example,
560 are a few of the more popular Unix flavors:
562 uname $^O $Config{'archname'}
563 --------------------------------------------
565 BSD/OS bsdos i386-bsdos
566 dgux dgux AViiON-dgux
567 DYNIX/ptx dynixptx i386-dynixptx
568 FreeBSD freebsd freebsd-i386
569 Linux linux arm-linux
570 Linux linux i386-linux
571 Linux linux i586-linux
572 Linux linux ppc-linux
573 HP-UX hpux PA-RISC1.1
575 Mac OS X rhapsody rhapsody
576 MachTen PPC machten powerpc-machten
578 NeXT 4 next OPENSTEP-Mach
579 openbsd openbsd i386-openbsd
580 OSF1 dec_osf alpha-dec_osf
581 reliantunix-n svr4 RM400-svr4
582 SCO_SV sco_sv i386-sco_sv
583 SINIX-N svr4 RM400-svr4
584 sn4609 unicos CRAY_C90-unicos
585 sn6521 unicosmk t3e-unicosmk
586 sn9617 unicos CRAY_J90-unicos
587 SunOS solaris sun4-solaris
588 SunOS solaris i86pc-solaris
589 SunOS4 sunos sun4-sunos
591 Because the value of C<$Config{archname}> may depend on the
592 hardware architecture, it can vary more than the value of C<$^O>.
594 =head2 DOS and Derivatives
596 Perl has long been ported to Intel-style microcomputers running under
597 systems like PC-DOS, MS-DOS, OS/2, and most Windows platforms you can
598 bring yourself to mention (except for Windows CE, if you count that).
599 Users familiar with I<COMMAND.COM> or I<CMD.EXE> style shells should
600 be aware that each of these file specifications may have subtle
603 $filespec0 = "c:/foo/bar/file.txt";
604 $filespec1 = "c:\\foo\\bar\\file.txt";
605 $filespec2 = 'c:\foo\bar\file.txt';
606 $filespec3 = 'c:\\foo\\bar\\file.txt';
608 System calls accept either C</> or C<\> as the path separator.
609 However, many command-line utilities of DOS vintage treat C</> as
610 the option prefix, so may get confused by filenames containing C</>.
611 Aside from calling any external programs, C</> will work just fine,
612 and probably better, as it is more consistent with popular usage,
613 and avoids the problem of remembering what to backwhack and what
616 The DOS FAT filesystem can accommodate only "8.3" style filenames. Under
617 the "case-insensitive, but case-preserving" HPFS (OS/2) and NTFS (NT)
618 filesystems you may have to be careful about case returned with functions
619 like C<readdir> or used with functions like C<open> or C<opendir>.
621 DOS also treats several filenames as special, such as AUX, PRN,
622 NUL, CON, COM1, LPT1, LPT2, etc. Unfortunately, sometimes these
623 filenames won't even work if you include an explicit directory
624 prefix. It is best to avoid such filenames, if you want your code
625 to be portable to DOS and its derivatives. It's hard to know what
626 these all are, unfortunately.
628 Users of these operating systems may also wish to make use of
629 scripts such as I<pl2bat.bat> or I<pl2cmd> to
630 put wrappers around your scripts.
632 Newline (C<\n>) is translated as C<\015\012> by STDIO when reading from
633 and writing to files (see L<"Newlines">). C<binmode(FILEHANDLE)>
634 will keep C<\n> translated as C<\012> for that filehandle. Since it is a
635 no-op on other systems, C<binmode> should be used for cross-platform code
636 that deals with binary data. That's assuming you realize in advance
637 that your data is in binary. General-purpose programs should
638 often assume nothing about their data.
640 The C<$^O> variable and the C<$Config{archname}> values for various
641 DOSish perls are as follows:
643 OS $^O $Config{'archname'}
644 --------------------------------------------
648 Windows 95 MSWin32 MSWin32-x86
649 Windows 98 MSWin32 MSWin32-x86
650 Windows NT MSWin32 MSWin32-x86
651 Windows NT MSWin32 MSWin32-ALPHA
652 Windows NT MSWin32 MSWin32-ppc
661 The djgpp environment for DOS, http://www.delorie.com/djgpp/
666 The EMX environment for DOS, OS/2, etc. emx@iaehv.nl,
667 http://www.leo.org/pub/comp/os/os2/leo/gnu/emx+gcc/index.html or
668 ftp://hobbes.nmsu.edu/pub/os2/dev/emx. Also L<perlos2>.
672 Build instructions for Win32 in L<perlwin32>, or under the Cygnus environment
677 The C<Win32::*> modules in L<Win32>.
681 The ActiveState Pages, http://www.activestate.com/
685 The Cygwin environment for Win32; F<README.cygwin> (installed
686 as L<perlcygwin>), http://sources.redhat.com/cygwin/
690 The U/WIN environment for Win32,
691 <http://www.research.att.com/sw/tools/uwin/
693 =item Build instructions for OS/2, L<perlos2>
700 Any module requiring XS compilation is right out for most people, because
701 MacPerl is built using non-free (and non-cheap!) compilers. Some XS
702 modules that can work with MacPerl are built and distributed in binary
705 Directories are specified as:
707 volume:folder:file for absolute pathnames
708 volume:folder: for absolute pathnames
709 :folder:file for relative pathnames
710 :folder: for relative pathnames
711 :file for relative pathnames
712 file for relative pathnames
714 Files are stored in the directory in alphabetical order. Filenames are
715 limited to 31 characters, and may include any character except for
716 null and C<:>, which is reserved as the path separator.
718 Instead of C<flock>, see C<FSpSetFLock> and C<FSpRstFLock> in the
719 Mac::Files module, or C<chmod(0444, ...)> and C<chmod(0666, ...)>.
721 In the MacPerl application, you can't run a program from the command line;
722 programs that expect C<@ARGV> to be populated can be edited with something
723 like the following, which brings up a dialog box asking for the command
727 @ARGV = split /\s+/, MacPerl::Ask('Arguments?');
730 A MacPerl script saved as a "droplet" will populate C<@ARGV> with the full
731 pathnames of the files dropped onto the script.
733 Mac users can run programs under a type of command line interface
734 under MPW (Macintosh Programmer's Workshop, a free development
735 environment from Apple). MacPerl was first introduced as an MPW
736 tool, and MPW can be used like a shell:
738 perl myscript.plx some arguments
740 ToolServer is another app from Apple that provides access to MPW tools
741 from MPW and the MacPerl app, which allows MacPerl programs to use
742 C<system>, backticks, and piped C<open>.
744 "S<Mac OS>" is the proper name for the operating system, but the value
745 in C<$^O> is "MacOS". To determine architecture, version, or whether
746 the application or MPW tool version is running, check:
748 $is_app = $MacPerl::Version =~ /App/;
749 $is_tool = $MacPerl::Version =~ /MPW/;
750 ($version) = $MacPerl::Version =~ /^(\S+)/;
751 $is_ppc = $MacPerl::Architecture eq 'MacPPC';
752 $is_68k = $MacPerl::Architecture eq 'Mac68K';
754 S<Mac OS X> and S<Mac OS X Server>, based on NeXT's OpenStep OS, will
755 (in theory) be able to run MacPerl natively, under the "Classic"
756 environment. The new "Cocoa" environment (formerly called the "Yellow Box")
757 may run a slightly modified version of MacPerl, using the Carbon interfaces.
759 S<Mac OS X Server> and its Open Source version, Darwin, both run Unix
760 perl natively (with a few patches). Full support for these
761 is slated for perl 5.6.
769 The MacPerl Pages, http://www.macperl.com/ .
773 The MacPerl mailing lists, http://www.macperl.org/ .
777 MacPerl Module Porters, http://pudge.net/mmp/ .
783 Perl on VMS is discussed in L<perlvms> in the perl distribution.
784 Perl on VMS can accept either VMS- or Unix-style file
785 specifications as in either of the following:
787 $ perl -ne "print if /perl_setup/i" SYS$LOGIN:LOGIN.COM
788 $ perl -ne "print if /perl_setup/i" /sys$login/login.com
790 but not a mixture of both as in:
792 $ perl -ne "print if /perl_setup/i" sys$login:/login.com
793 Can't open sys$login:/login.com: file specification syntax error
795 Interacting with Perl from the Digital Command Language (DCL) shell
796 often requires a different set of quotation marks than Unix shells do.
799 $ perl -e "print ""Hello, world.\n"""
802 There are several ways to wrap your perl scripts in DCL F<.COM> files, if
803 you are so inclined. For example:
805 $ write sys$output "Hello from DCL!"
807 $ then perl -x 'f$environment("PROCEDURE")
808 $ else perl -x - 'p1 'p2 'p3 'p4 'p5 'p6 'p7 'p8
809 $ deck/dollars="__END__"
812 print "Hello from Perl!\n";
817 Do take care with C<$ ASSIGN/nolog/user SYS$COMMAND: SYS$INPUT> if your
818 perl-in-DCL script expects to do things like C<< $read = <STDIN>; >>.
820 Filenames are in the format "name.extension;version". The maximum
821 length for filenames is 39 characters, and the maximum length for
822 extensions is also 39 characters. Version is a number from 1 to
823 32767. Valid characters are C</[A-Z0-9$_-]/>.
825 VMS's RMS filesystem is case-insensitive and does not preserve case.
826 C<readdir> returns lowercased filenames, but specifying a file for
827 opening remains case-insensitive. Files without extensions have a
828 trailing period on them, so doing a C<readdir> with a file named F<A.;5>
829 will return F<a.> (though that file could be opened with
832 RMS had an eight level limit on directory depths from any rooted logical
833 (allowing 16 levels overall) prior to VMS 7.2. Hence
834 C<PERL_ROOT:[LIB.2.3.4.5.6.7.8]> is a valid directory specification but
835 C<PERL_ROOT:[LIB.2.3.4.5.6.7.8.9]> is not. F<Makefile.PL> authors might
836 have to take this into account, but at least they can refer to the former
837 as C</PERL_ROOT/lib/2/3/4/5/6/7/8/>.
839 The VMS::Filespec module, which gets installed as part of the build
840 process on VMS, is a pure Perl module that can easily be installed on
841 non-VMS platforms and can be helpful for conversions to and from RMS
844 What C<\n> represents depends on the type of file opened. It could
845 be C<\015>, C<\012>, C<\015\012>, or nothing. The VMS::Stdio module
846 provides access to the special fopen() requirements of files with unusual
849 TCP/IP stacks are optional on VMS, so socket routines might not be
850 implemented. UDP sockets may not be supported.
852 The value of C<$^O> on OpenVMS is "VMS". To determine the architecture
853 that you are running on without resorting to loading all of C<%Config>
854 you can examine the content of the C<@INC> array like so:
856 if (grep(/VMS_AXP/, @INC)) {
857 print "I'm on Alpha!\n";
859 } elsif (grep(/VMS_VAX/, @INC)) {
860 print "I'm on VAX!\n";
863 print "I'm not so sure about where $^O is...\n";
866 On VMS, perl determines the UTC offset from the C<SYS$TIMEZONE_DIFFERENTIAL>
867 logical name. Although the VMS epoch began at 17-NOV-1858 00:00:00.00,
868 calls to C<localtime> are adjusted to count offsets from
869 01-JAN-1970 00:00:00.00, just like Unix.
877 F<README.vms> (installed as L<README_vms>), L<perlvms>
881 vmsperl list, majordomo@perl.org
883 (Put the words C<subscribe vmsperl> in message body.)
887 vmsperl on the web, http://www.sidhe.org/vmsperl/index.html
893 Perl on VOS is discussed in F<README.vos> in the perl distribution.
894 Perl on VOS can accept either VOS- or Unix-style file
895 specifications as in either of the following:
897 $ perl -ne "print if /perl_setup/i" >system>notices
898 $ perl -ne "print if /perl_setup/i" /system/notices
900 or even a mixture of both as in:
902 $ perl -ne "print if /perl_setup/i" >system/notices
904 Even though VOS allows the slash character to appear in object
905 names, because the VOS port of Perl interprets it as a pathname
906 delimiting character, VOS files, directories, or links whose names
907 contain a slash character cannot be processed. Such files must be
908 renamed before they can be processed by Perl. Note that VOS limits
909 file names to 32 or fewer characters.
911 The following C functions are unimplemented on VOS, and any attempt by
912 Perl to use them will result in a fatal error message and an immediate
913 exit from Perl: dup, do_aspawn, do_spawn, fork, waitpid. Once these
914 functions become available in the VOS POSIX.1 implementation, you can
915 either recompile and rebind Perl, or you can download a newer port from
918 The value of C<$^O> on VOS is "VOS". To determine the architecture that
919 you are running on without resorting to loading all of C<%Config> you
920 can examine the content of the @INC array like so:
923 print "I'm on a Stratus box!\n";
925 print "I'm not on a Stratus box!\n";
929 if (grep(/860/, @INC)) {
930 print "This box is a Stratus XA/R!\n";
932 } elsif (grep(/7100/, @INC)) {
933 print "This box is a Stratus HP 7100 or 8xxx!\n";
935 } elsif (grep(/8000/, @INC)) {
936 print "This box is a Stratus HP 8xxx!\n";
939 print "This box is a Stratus 68K!\n";
952 The VOS mailing list.
954 There is no specific mailing list for Perl on VOS. You can post
955 comments to the comp.sys.stratus newsgroup, or subscribe to the general
956 Stratus mailing list. Send a letter with "Subscribe Info-Stratus" in
957 the message body to majordomo@list.stratagy.com.
961 VOS Perl on the web at http://ftp.stratus.com/pub/vos/vos.html
965 =head2 EBCDIC Platforms
967 Recent versions of Perl have been ported to platforms such as OS/400 on
968 AS/400 minicomputers as well as OS/390, VM/ESA, and BS2000 for S/390
969 Mainframes. Such computers use EBCDIC character sets internally (usually
970 Character Code Set ID 0037 for OS/400 and either 1047 or POSIX-BC for S/390
971 systems). On the mainframe perl currently works under the "Unix system
972 services for OS/390" (formerly known as OpenEdition), VM/ESA OpenEdition, or
973 the BS200 POSIX-BC system (BS2000 is supported in perl 5.6 and greater).
974 See L<perlos390> for details.
976 As of R2.5 of USS for OS/390 and Version 2.3 of VM/ESA these Unix
977 sub-systems do not support the C<#!> shebang trick for script invocation.
978 Hence, on OS/390 and VM/ESA perl scripts can be executed with a header
979 similar to the following simple script:
982 eval 'exec /usr/local/bin/perl -S $0 ${1+"$@"}'
984 #!/usr/local/bin/perl # just a comment really
986 print "Hello from perl!\n";
988 OS/390 will support the C<#!> shebang trick in release 2.8 and beyond.
989 Calls to C<system> and backticks can use POSIX shell syntax on all
992 On the AS/400, if PERL5 is in your library list, you may need
993 to wrap your perl scripts in a CL procedure to invoke them like so:
996 CALL PGM(PERL5/PERL) PARM('/QOpenSys/hello.pl')
999 This will invoke the perl script F<hello.pl> in the root of the
1000 QOpenSys file system. On the AS/400 calls to C<system> or backticks
1003 On these platforms, bear in mind that the EBCDIC character set may have
1004 an effect on what happens with some perl functions (such as C<chr>,
1005 C<pack>, C<print>, C<printf>, C<ord>, C<sort>, C<sprintf>, C<unpack>), as
1006 well as bit-fiddling with ASCII constants using operators like C<^>, C<&>
1007 and C<|>, not to mention dealing with socket interfaces to ASCII computers
1008 (see L<"Newlines">).
1010 Fortunately, most web servers for the mainframe will correctly
1011 translate the C<\n> in the following statement to its ASCII equivalent
1012 (C<\r> is the same under both Unix and OS/390 & VM/ESA):
1014 print "Content-type: text/html\r\n\r\n";
1016 The values of C<$^O> on some of these platforms includes:
1018 uname $^O $Config{'archname'}
1019 --------------------------------------------
1022 POSIX-BC posix-bc BS2000-posix-bc
1025 Some simple tricks for determining if you are running on an EBCDIC
1026 platform could include any of the following (perhaps all):
1028 if ("\t" eq "\05") { print "EBCDIC may be spoken here!\n"; }
1030 if (ord('A') == 193) { print "EBCDIC may be spoken here!\n"; }
1032 if (chr(169) eq 'z') { print "EBCDIC may be spoken here!\n"; }
1034 One thing you may not want to rely on is the EBCDIC encoding
1035 of punctuation characters since these may differ from code page to code
1036 page (and once your module or script is rumoured to work with EBCDIC,
1037 folks will want it to work with all EBCDIC character sets).
1047 L<perlos390>, F<README.os390>, F<perlposix-bc>, F<README.vmesa>,
1052 The perl-mvs@perl.org list is for discussion of porting issues as well as
1053 general usage issues for all EBCDIC Perls. Send a message body of
1054 "subscribe perl-mvs" to majordomo@perl.org.
1058 AS/400 Perl information at
1059 http://as400.rochester.ibm.com/
1060 as well as on CPAN in the F<ports/> directory.
1064 =head2 Acorn RISC OS
1066 Because Acorns use ASCII with newlines (C<\n>) in text files as C<\012> like
1067 Unix, and because Unix filename emulation is turned on by default,
1068 most simple scripts will probably work "out of the box". The native
1069 filesystem is modular, and individual filesystems are free to be
1070 case-sensitive or insensitive, and are usually case-preserving. Some
1071 native filesystems have name length limits, which file and directory
1072 names are silently truncated to fit. Scripts should be aware that the
1073 standard filesystem currently has a name length limit of B<10>
1074 characters, with up to 77 items in a directory, but other filesystems
1075 may not impose such limitations.
1077 Native filenames are of the form
1079 Filesystem#Special_Field::DiskName.$.Directory.Directory.File
1083 Special_Field is not usually present, but may contain . and $ .
1084 Filesystem =~ m|[A-Za-z0-9_]|
1085 DsicName =~ m|[A-Za-z0-9_/]|
1086 $ represents the root directory
1087 . is the path separator
1088 @ is the current directory (per filesystem but machine global)
1089 ^ is the parent directory
1090 Directory and File =~ m|[^\0- "\.\$\%\&:\@\\^\|\177]+|
1092 The default filename translation is roughly C<tr|/.|./|;>
1094 Note that C<"ADFS::HardDisk.$.File" ne 'ADFS::HardDisk.$.File'> and that
1095 the second stage of C<$> interpolation in regular expressions will fall
1096 foul of the C<$.> if scripts are not careful.
1098 Logical paths specified by system variables containing comma-separated
1099 search lists are also allowed; hence C<System:Modules> is a valid
1100 filename, and the filesystem will prefix C<Modules> with each section of
1101 C<System$Path> until a name is made that points to an object on disk.
1102 Writing to a new file C<System:Modules> would be allowed only if
1103 C<System$Path> contains a single item list. The filesystem will also
1104 expand system variables in filenames if enclosed in angle brackets, so
1105 C<< <System$Dir>.Modules >> would look for the file
1106 S<C<$ENV{'System$Dir'} . 'Modules'>>. The obvious implication of this is
1107 that B<fully qualified filenames can start with C<< <> >>> and should
1108 be protected when C<open> is used for input.
1110 Because C<.> was in use as a directory separator and filenames could not
1111 be assumed to be unique after 10 characters, Acorn implemented the C
1112 compiler to strip the trailing C<.c> C<.h> C<.s> and C<.o> suffix from
1113 filenames specified in source code and store the respective files in
1114 subdirectories named after the suffix. Hence files are translated:
1117 C:foo.h C:h.foo (logical path variable)
1118 sys/os.h sys.h.os (C compiler groks Unix-speak)
1119 10charname.c c.10charname
1120 10charname.o o.10charname
1121 11charname_.c c.11charname (assuming filesystem truncates at 10)
1123 The Unix emulation library's translation of filenames to native assumes
1124 that this sort of translation is required, and it allows a user-defined list
1125 of known suffixes that it will transpose in this fashion. This may
1126 seem transparent, but consider that with these rules C<foo/bar/baz.h>
1127 and C<foo/bar/h/baz> both map to C<foo.bar.h.baz>, and that C<readdir> and
1128 C<glob> cannot and do not attempt to emulate the reverse mapping. Other
1129 C<.>'s in filenames are translated to C</>.
1131 As implied above, the environment accessed through C<%ENV> is global, and
1132 the convention is that program specific environment variables are of the
1133 form C<Program$Name>. Each filesystem maintains a current directory,
1134 and the current filesystem's current directory is the B<global> current
1135 directory. Consequently, sociable programs don't change the current
1136 directory but rely on full pathnames, and programs (and Makefiles) cannot
1137 assume that they can spawn a child process which can change the current
1138 directory without affecting its parent (and everyone else for that
1141 Because native operating system filehandles are global and are currently
1142 allocated down from 255, with 0 being a reserved value, the Unix emulation
1143 library emulates Unix filehandles. Consequently, you can't rely on
1144 passing C<STDIN>, C<STDOUT>, or C<STDERR> to your children.
1146 The desire of users to express filenames of the form
1147 C<< <Foo$Dir>.Bar >> on the command line unquoted causes problems,
1148 too: C<``> command output capture has to perform a guessing game. It
1149 assumes that a string C<< <[^<>]+\$[^<>]> >> is a
1150 reference to an environment variable, whereas anything else involving
1151 C<< < >> or C<< > >> is redirection, and generally manages to be 99%
1152 right. Of course, the problem remains that scripts cannot rely on any
1153 Unix tools being available, or that any tools found have Unix-like command
1156 Extensions and XS are, in theory, buildable by anyone using free
1157 tools. In practice, many don't, as users of the Acorn platform are
1158 used to binary distributions. MakeMaker does run, but no available
1159 make currently copes with MakeMaker's makefiles; even if and when
1160 this should be fixed, the lack of a Unix-like shell will cause
1161 problems with makefile rules, especially lines of the form C<cd
1162 sdbm && make all>, and anything using quoting.
1164 "S<RISC OS>" is the proper name for the operating system, but the value
1165 in C<$^O> is "riscos" (because we don't like shouting).
1169 Perl has been ported to many platforms that do not fit into any of
1170 the categories listed above. Some, such as AmigaOS, Atari MiNT,
1171 BeOS, HP MPE/iX, QNX, Plan 9, and VOS, have been well-integrated
1172 into the standard Perl source code kit. You may need to see the
1173 F<ports/> directory on CPAN for information, and possibly binaries,
1174 for the likes of: aos, Atari ST, lynxos, riscos, Novell Netware,
1175 Tandem Guardian, I<etc.> (Yes, we know that some of these OSes may
1176 fall under the Unix category, but we are not a standards body.)
1178 Some approximate operating system names and their C<$^O> values
1179 in the "OTHER" category include:
1181 OS $^O $Config{'archname'}
1182 ------------------------------------------
1183 Amiga DOS amigaos m68k-amigos
1184 MPE/iX mpeix PA-RISC1.1
1192 Amiga, F<README.amiga> (installed as L<perlamiga>).
1196 Atari, F<README.mint> and Guido Flohr's web page
1197 http://stud.uni-sb.de/~gufl0000/
1201 Be OS, F<README.beos>
1205 HP 300 MPE/iX, F<README.mpeix> and Mark Bixby's web page
1206 http://www.cccd.edu/~markb/perlix.html
1210 A free perl5-based PERL.NLM for Novell Netware is available in
1211 precompiled binary and source code form from http://www.novell.com/
1212 as well as from CPAN.
1216 Plan 9, F<README.plan9>
1220 =head1 FUNCTION IMPLEMENTATIONS
1222 Listed below are functions that are either completely unimplemented
1223 or else have been implemented differently on various platforms.
1224 Following each description will be, in parentheses, a list of
1225 platforms that the description applies to.
1227 The list may well be incomplete, or even wrong in some places. When
1228 in doubt, consult the platform-specific README files in the Perl
1229 source distribution, and any other documentation resources accompanying
1232 Be aware, moreover, that even among Unix-ish systems there are variations.
1234 For many functions, you can also query C<%Config>, exported by
1235 default from the Config module. For example, to check whether the
1236 platform has the C<lstat> call, check C<$Config{d_lstat}>. See
1237 L<Config> for a full description of available variables.
1239 =head2 Alphabetical Listing of Perl Functions
1249 C<-r>, C<-w>, and C<-x> have a limited meaning only; directories
1250 and applications are executable, and there are no uid/gid
1251 considerations. C<-o> is not supported. (S<Mac OS>)
1253 C<-r>, C<-w>, C<-x>, and C<-o> tell whether the file is accessible,
1254 which may not reflect UIC-based file protections. (VMS)
1256 C<-s> returns the size of the data fork, not the total size of data fork
1257 plus resource fork. (S<Mac OS>).
1259 C<-s> by name on an open file will return the space reserved on disk,
1260 rather than the current extent. C<-s> on an open filehandle returns the
1261 current size. (S<RISC OS>)
1263 C<-R>, C<-W>, C<-X>, C<-O> are indistinguishable from C<-r>, C<-w>,
1264 C<-x>, C<-o>. (S<Mac OS>, Win32, VMS, S<RISC OS>)
1266 C<-b>, C<-c>, C<-k>, C<-g>, C<-p>, C<-u>, C<-A> are not implemented.
1269 C<-g>, C<-k>, C<-l>, C<-p>, C<-u>, C<-A> are not particularly meaningful.
1270 (Win32, VMS, S<RISC OS>)
1272 C<-d> is true if passed a device spec without an explicit directory.
1275 C<-T> and C<-B> are implemented, but might misclassify Mac text files
1276 with foreign characters; this is the case will all platforms, but may
1277 affect S<Mac OS> often. (S<Mac OS>)
1279 C<-x> (or C<-X>) determine if a file ends in one of the executable
1280 suffixes. C<-S> is meaningless. (Win32)
1282 C<-x> (or C<-X>) determine if a file has an executable file type.
1289 Not implemented. (Win32)
1291 =item binmode FILEHANDLE
1293 Meaningless. (S<Mac OS>, S<RISC OS>)
1295 Reopens file and restores pointer; if function fails, underlying
1296 filehandle may be closed, or pointer may be in a different position.
1299 The value returned by C<tell> may be affected after the call, and
1300 the filehandle may be flushed. (Win32)
1304 Only limited meaning. Disabling/enabling write permission is mapped to
1305 locking/unlocking the file. (S<Mac OS>)
1307 Only good for changing "owner" read-write access, "group", and "other"
1308 bits are meaningless. (Win32)
1310 Only good for changing "owner" and "other" read-write access. (S<RISC OS>)
1312 Access permissions are mapped onto VOS access-control list changes. (VOS)
1316 Not implemented. (S<Mac OS>, Win32, Plan9, S<RISC OS>, VOS)
1318 Does nothing, but won't fail. (Win32)
1320 =item chroot FILENAME
1324 Not implemented. (S<Mac OS>, Win32, VMS, Plan9, S<RISC OS>, VOS, VM/ESA)
1326 =item crypt PLAINTEXT,SALT
1328 May not be available if library or source was not provided when building
1331 Not implemented. (VOS)
1335 Not implemented. (VMS, Plan9, VOS)
1337 =item dbmopen HASH,DBNAME,MODE
1339 Not implemented. (VMS, Plan9, VOS)
1343 Not useful. (S<Mac OS>, S<RISC OS>)
1345 Not implemented. (Win32)
1347 Invokes VMS debugger. (VMS)
1351 Not implemented. (S<Mac OS>)
1353 Implemented via Spawn. (VM/ESA)
1355 Does not automatically flush output handles on some platforms.
1356 (SunOS, Solaris, HP-UX)
1358 =item fcntl FILEHANDLE,FUNCTION,SCALAR
1360 Not implemented. (Win32, VMS)
1362 =item flock FILEHANDLE,OPERATION
1364 Not implemented (S<Mac OS>, VMS, S<RISC OS>, VOS).
1366 Available only on Windows NT (not on Windows 95). (Win32)
1370 Not implemented. (S<Mac OS>, AmigaOS, S<RISC OS>, VOS, VM/ESA)
1372 Emulated using multiple interpreters. See L<perlfork>. (Win32)
1374 Does not automatically flush output handles on some platforms.
1375 (SunOS, Solaris, HP-UX)
1379 Not implemented. (S<Mac OS>, S<RISC OS>)
1383 Not implemented. (S<Mac OS>, Win32, VMS, S<RISC OS>, VOS)
1387 Not implemented. (S<Mac OS>, Win32, VMS, S<RISC OS>)
1389 =item getpriority WHICH,WHO
1391 Not implemented. (S<Mac OS>, Win32, VMS, S<RISC OS>, VOS, VM/ESA)
1395 Not implemented. (S<Mac OS>, Win32)
1397 Not useful. (S<RISC OS>)
1401 Not implemented. (S<Mac OS>, Win32, VMS, S<RISC OS>)
1403 =item getnetbyname NAME
1405 Not implemented. (S<Mac OS>, Win32, Plan9)
1409 Not implemented. (S<Mac OS>, Win32)
1411 Not useful. (S<RISC OS>)
1415 Not implemented. (S<Mac OS>, Win32, VMS, S<RISC OS>)
1417 =item getnetbyaddr ADDR,ADDRTYPE
1419 Not implemented. (S<Mac OS>, Win32, Plan9)
1421 =item getprotobynumber NUMBER
1423 Not implemented. (S<Mac OS>)
1425 =item getservbyport PORT,PROTO
1427 Not implemented. (S<Mac OS>)
1431 Not implemented. (S<Mac OS>, Win32, VM/ESA)
1435 Not implemented. (S<Mac OS>, Win32, VMS, VM/ESA)
1439 Not implemented. (S<Mac OS>, Win32)
1443 Not implemented. (S<Mac OS>, Win32, Plan9)
1447 Not implemented. (S<Mac OS>, Win32, Plan9)
1451 Not implemented. (Win32, Plan9)
1455 Not implemented. (S<Mac OS>, Win32, S<RISC OS>)
1459 Not implemented. (S<Mac OS>, Win32, VMS, S<RISC OS>)
1461 =item sethostent STAYOPEN
1463 Not implemented. (S<Mac OS>, Win32, Plan9, S<RISC OS>)
1465 =item setnetent STAYOPEN
1467 Not implemented. (S<Mac OS>, Win32, Plan9, S<RISC OS>)
1469 =item setprotoent STAYOPEN
1471 Not implemented. (S<Mac OS>, Win32, Plan9, S<RISC OS>)
1473 =item setservent STAYOPEN
1475 Not implemented. (Plan9, Win32, S<RISC OS>)
1479 Not implemented. (S<Mac OS>, MPE/iX, VM/ESA, Win32)
1483 Not implemented. (S<Mac OS>, MPE/iX, S<RISC OS>, VM/ESA, VMS, Win32)
1487 Not implemented. (S<Mac OS>, Win32)
1491 Not implemented. (S<Mac OS>, Win32, Plan9)
1495 Not implemented. (S<Mac OS>, Win32, Plan9)
1499 Not implemented. (Plan9, Win32)
1501 =item getsockopt SOCKET,LEVEL,OPTNAME
1503 Not implemented. (S<Mac OS>, Plan9)
1509 Globbing built-in, but only C<*> and C<?> metacharacters are supported.
1512 This operator is implemented via the File::Glob extension on most
1513 platforms. See L<File::Glob> for portability information.
1515 =item ioctl FILEHANDLE,FUNCTION,SCALAR
1517 Not implemented. (VMS)
1519 Available only for socket handles, and it does what the ioctlsocket() call
1520 in the Winsock API does. (Win32)
1522 Available only for socket handles. (S<RISC OS>)
1524 =item kill SIGNAL, LIST
1526 Not implemented, hence not useful for taint checking. (S<Mac OS>,
1529 C<kill()> doesn't have the semantics of C<raise()>, i.e. it doesn't send
1530 a signal to the identified process like it does on Unix platforms.
1531 Instead C<kill($sig, $pid)> terminates the process identified by $pid,
1532 and makes it exit immediately with exit status $sig. As in Unix, if
1533 $sig is 0 and the specified process exists, it returns true without
1534 actually terminating it. (Win32)
1536 =item link OLDFILE,NEWFILE
1538 Not implemented. (S<Mac OS>, MPE/iX, VMS, S<RISC OS>)
1540 Link count not updated because hard links are not quite that hard
1541 (They are sort of half-way between hard and soft links). (AmigaOS)
1543 Hard links are implemented on Win32 (Windows NT and Windows 2000)
1546 =item lstat FILEHANDLE
1552 Not implemented. (VMS, S<RISC OS>)
1554 Return values (especially for device and inode) may be bogus. (Win32)
1556 =item msgctl ID,CMD,ARG
1558 =item msgget KEY,FLAGS
1560 =item msgsnd ID,MSG,FLAGS
1562 =item msgrcv ID,VAR,SIZE,TYPE,FLAGS
1564 Not implemented. (S<Mac OS>, Win32, VMS, Plan9, S<RISC OS>, VOS)
1566 =item open FILEHANDLE,EXPR
1568 =item open FILEHANDLE
1570 The C<|> variants are supported only if ToolServer is installed.
1573 open to C<|-> and C<-|> are unsupported. (S<Mac OS>, Win32, S<RISC OS>)
1575 Opening a process does not automatically flush output handles on some
1576 platforms. (SunOS, Solaris, HP-UX)
1578 =item pipe READHANDLE,WRITEHANDLE
1580 Not implemented. (S<Mac OS>)
1582 Very limited functionality. (MiNT)
1588 Not implemented. (Win32, VMS, S<RISC OS>)
1590 =item select RBITS,WBITS,EBITS,TIMEOUT
1592 Only implemented on sockets. (Win32)
1594 Only reliable on sockets. (S<RISC OS>)
1596 Note that the C<socket FILEHANDLE> form is generally portable.
1598 =item semctl ID,SEMNUM,CMD,ARG
1600 =item semget KEY,NSEMS,FLAGS
1602 =item semop KEY,OPSTRING
1604 Not implemented. (S<Mac OS>, Win32, VMS, S<RISC OS>, VOS)
1608 Not implemented. (MPE/iX, Win32)
1610 =item setpgrp PID,PGRP
1612 Not implemented. (S<Mac OS>, Win32, VMS, S<RISC OS>, VOS)
1614 =item setpriority WHICH,WHO,PRIORITY
1616 Not implemented. (S<Mac OS>, Win32, VMS, S<RISC OS>, VOS)
1620 Not implemented. (MPE/iX, Win32)
1622 =item setsockopt SOCKET,LEVEL,OPTNAME,OPTVAL
1624 Not implemented. (S<Mac OS>, Plan9)
1626 =item shmctl ID,CMD,ARG
1628 =item shmget KEY,SIZE,FLAGS
1630 =item shmread ID,VAR,POS,SIZE
1632 =item shmwrite ID,STRING,POS,SIZE
1634 Not implemented. (S<Mac OS>, Win32, VMS, S<RISC OS>, VOS)
1636 =item socketpair SOCKET1,SOCKET2,DOMAIN,TYPE,PROTOCOL
1638 Not implemented. (S<Mac OS>, Win32, VMS, S<RISC OS>, VOS, VM/ESA)
1640 =item stat FILEHANDLE
1646 Platforms that do not have rdev, blksize, or blocks will return these
1647 as '', so numeric comparison or manipulation of these fields may cause
1648 'not numeric' warnings.
1650 mtime and atime are the same thing, and ctime is creation time instead of
1651 inode change time. (S<Mac OS>)
1653 device and inode are not meaningful. (Win32)
1655 device and inode are not necessarily reliable. (VMS)
1657 mtime, atime and ctime all return the last modification time. Device and
1658 inode are not necessarily reliable. (S<RISC OS>)
1660 dev, rdev, blksize, and blocks are not available. inode is not
1661 meaningful and will differ between stat calls on the same file. (os2)
1663 =item symlink OLDFILE,NEWFILE
1665 Not implemented. (Win32, VMS, S<RISC OS>)
1669 Not implemented. (S<Mac OS>, Win32, VMS, S<RISC OS>, VOS, VM/ESA)
1671 =item sysopen FILEHANDLE,FILENAME,MODE,PERMS
1673 The traditional "0", "1", and "2" MODEs are implemented with different
1674 numeric values on some systems. The flags exported by C<Fcntl>
1675 (O_RDONLY, O_WRONLY, O_RDWR) should work everywhere though. (S<Mac
1676 OS>, OS/390, VM/ESA)
1680 Only implemented if ToolServer is installed. (S<Mac OS>)
1682 As an optimization, may not call the command shell specified in
1683 C<$ENV{PERL5SHELL}>. C<system(1, @args)> spawns an external
1684 process and immediately returns its process designator, without
1685 waiting for it to terminate. Return value may be used subsequently
1686 in C<wait> or C<waitpid>. Failure to spawn() a subprocess is indicated
1687 by setting $? to "255 << 8". C<$?> is set in a way compatible with
1688 Unix (i.e. the exitstatus of the subprocess is obtained by "$? >> 8",
1689 as described in the documentation). (Win32)
1691 There is no shell to process metacharacters, and the native standard is
1692 to pass a command line terminated by "\n" "\r" or "\0" to the spawned
1693 program. Redirection such as C<< > foo >> is performed (if at all) by
1694 the run time library of the spawned program. C<system> I<list> will call
1695 the Unix emulation library's C<exec> emulation, which attempts to provide
1696 emulation of the stdin, stdout, stderr in force in the parent, providing
1697 the child program uses a compatible version of the emulation library.
1698 I<scalar> will call the native command line direct and no such emulation
1699 of a child Unix program will exists. Mileage B<will> vary. (S<RISC OS>)
1701 Far from being POSIX compliant. Because there may be no underlying
1702 /bin/sh tries to work around the problem by forking and execing the
1703 first token in its argument string. Handles basic redirection
1704 ("<" or ">") on its own behalf. (MiNT)
1706 Does not automatically flush output handles on some platforms.
1707 (SunOS, Solaris, HP-UX)
1711 Only the first entry returned is nonzero. (S<Mac OS>)
1713 "cumulative" times will be bogus. On anything other than Windows NT
1714 or Windows 2000, "system" time will be bogus, and "user" time is
1715 actually the time returned by the clock() function in the C runtime
1718 Not useful. (S<RISC OS>)
1720 =item truncate FILEHANDLE,LENGTH
1722 =item truncate EXPR,LENGTH
1724 Not implemented. (VMS)
1726 Truncation to zero-length only. (VOS)
1728 If a FILEHANDLE is supplied, it must be writable and opened in append
1729 mode (i.e., use C<open(FH, '>>filename')>
1730 or C<sysopen(FH,...,O_APPEND|O_RDWR)>. If a filename is supplied, it
1731 should not be held open elsewhere. (Win32)
1737 Returns undef where unavailable, as of version 5.005.
1739 C<umask> works but the correct permissions are set only when the file
1740 is finally closed. (AmigaOS)
1744 Only the modification time is updated. (S<Mac OS>, VMS, S<RISC OS>)
1746 May not behave as expected. Behavior depends on the C runtime
1747 library's implementation of utime(), and the filesystem being
1748 used. The FAT filesystem typically does not support an "access
1749 time" field, and it may limit timestamps to a granularity of
1750 two seconds. (Win32)
1754 =item waitpid PID,FLAGS
1756 Not implemented. (S<Mac OS>, VOS)
1758 Can only be applied to process handles returned for processes spawned
1759 using C<system(1, ...)> or pseudo processes created with C<fork()>. (Win32)
1761 Not useful. (S<RISC OS>)
1769 =item v1.47, 22 March 2000
1771 Various cleanups from Tom Christiansen, including migration of
1772 long platform listings from L<perl>.
1774 =item v1.46, 12 February 2000
1776 Updates for VOS and MPE/iX. (Peter Prymmer) Other small changes.
1778 =item v1.45, 20 December 1999
1780 Small changes from 5.005_63 distribution, more changes to EBCDIC info.
1782 =item v1.44, 19 July 1999
1784 A bunch of updates from Peter Prymmer for C<$^O> values,
1785 endianness, File::Spec, VMS, BS2000, OS/400.
1787 =item v1.43, 24 May 1999
1789 Added a lot of cleaning up from Tom Christiansen.
1791 =item v1.42, 22 May 1999
1793 Added notes about tests, sprintf/printf, and epoch offsets.
1795 =item v1.41, 19 May 1999
1797 Lots more little changes to formatting and content.
1799 Added a bunch of C<$^O> and related values
1800 for various platforms; fixed mail and web addresses, and added
1801 and changed miscellaneous notes. (Peter Prymmer)
1803 =item v1.40, 11 April 1999
1805 Miscellaneous changes.
1807 =item v1.39, 11 February 1999
1809 Changes from Jarkko and EMX URL fixes Michael Schwern. Additional
1810 note about newlines added.
1812 =item v1.38, 31 December 1998
1814 More changes from Jarkko.
1816 =item v1.37, 19 December 1998
1818 More minor changes. Merge two separate version 1.35 documents.
1820 =item v1.36, 9 September 1998
1822 Updated for Stratus VOS. Also known as version 1.35.
1824 =item v1.35, 13 August 1998
1826 Integrate more minor changes, plus addition of new sections under
1827 L<"ISSUES">: L<"Numbers endianness and Width">,
1828 L<"Character sets and character encoding">,
1829 L<"Internationalisation">.
1831 =item v1.33, 06 August 1998
1833 Integrate more minor changes.
1835 =item v1.32, 05 August 1998
1837 Integrate more minor changes.
1839 =item v1.30, 03 August 1998
1841 Major update for RISC OS, other minor changes.
1843 =item v1.23, 10 July 1998
1845 First public release with perl5.005.
1849 =head1 Supported Platforms
1851 As of early March 2000 (the Perl release 5.6.0), the following
1852 platforms are able to build Perl from the standard source code
1853 distribution available at http://www.perl.com/CPAN/src/index.html
1885 1) in DOS mode either the DOS or OS/2 ports can be used
1886 2) new in 5.6.0: the BSD/NeXT-based UNIX of Mac OS X
1887 3) formerly known as Digital UNIX and before that DEC OSF/1
1888 4) compilers: Borland, Cygwin, Mingw32 EGCS/GCC, VC++
1890 The following platforms worked for the previous major release
1891 (5.005_03 being the latest maintenance release of that, as of early
1892 March 2000), but be did not manage to test these in time for the 5.6.0
1893 release of Perl. There is a very good chance that these will work
1894 just fine with 5.6.0.
1910 The following platform worked for the previous major release (5.005_03
1911 being the latest maintenance release of that, as of early March 2000).
1912 However, standardization on UTF-8 as the internal string representation
1913 in 5.6.0 has introduced incompatibilities in this EBCDIC platform.
1914 Support for this platform may be enabled in a future release:
1918 1) Previously known as MVS, or OpenEdition MVS.
1920 Strongly related to the OS390 platform by also being EBCDIC-based
1921 mainframe platforms are the following platforms:
1926 These are also not expected to work under 5.6.0 for the same reasons
1927 as OS390. Contact the mailing list perl-mvs@perl.org for more details.
1929 MacOS (Classic, pre-X) is almost 5.6.0-ready; building from the source
1930 does work with 5.6.0, but additional MacOS specific source code is needed
1931 for a complete port. Contact the mailing list macperl-porters@macperl.org
1932 for more information.
1934 The following platforms have been known to build Perl from source in
1935 the past, but we haven't been able to verify their status for the
1936 current release, either because the hardware/software platforms are
1937 rare or because we don't have an active champion on these
1969 Support for the following platform is planned for a future Perl release:
1973 The following platforms have their own source code distributions and
1974 binaries available via http://www.perl.com/CPAN/ports/index.html:
1980 Tandem Guardian 5.004
1982 The following platforms have only binaries available via
1983 http://www.perl.com/CPAN/ports/index.html :
1987 Acorn RISCOS 5.005_02
1991 Although we do suggest that you always build your own Perl from
1992 the source code, both for maximal configurability and for security,
1993 in case you are in a hurry you can check
1994 http://www.perl.com/CPAN/ports/index.html for binary distributions.
1998 L<perlaix>, L<perlamiga>, L<perlcygwin>, L<perldos>, L<perlebcdic>,
1999 L<perlhpux>, L<perlos2>, L<perlos390>, L<perlposix-bc>, L<perlwin32>,
2000 L<perlvms>, and L<Win32>.
2002 =head1 AUTHORS / CONTRIBUTORS
2004 Abigail <abigail@fnx.com>,
2005 Charles Bailey <bailey@newman.upenn.edu>,
2006 Graham Barr <gbarr@pobox.com>,
2007 Tom Christiansen <tchrist@perl.com>,
2008 Nicholas Clark <Nicholas.Clark@liverpool.ac.uk>,
2009 Thomas Dorner <Thomas.Dorner@start.de>,
2010 Andy Dougherty <doughera@lafcol.lafayette.edu>,
2011 Dominic Dunlop <domo@vo.lu>,
2012 Neale Ferguson <neale@mailbox.tabnsw.com.au>,
2013 David J. Fiander <davidf@mks.com>,
2014 Paul Green <Paul_Green@stratus.com>,
2015 M.J.T. Guy <mjtg@cus.cam.ac.uk>,
2016 Jarkko Hietaniemi <jhi@iki.fi>,
2017 Luther Huffman <lutherh@stratcom.com>,
2018 Nick Ing-Simmons <nick@ni-s.u-net.com>,
2019 Andreas J. KE<ouml>nig <koenig@kulturbox.de>,
2020 Markus Laker <mlaker@contax.co.uk>,
2021 Andrew M. Langmead <aml@world.std.com>,
2022 Larry Moore <ljmoore@freespace.net>,
2023 Paul Moore <Paul.Moore@uk.origin-it.com>,
2024 Chris Nandor <pudge@pobox.com>,
2025 Matthias Neeracher <neeri@iis.ee.ethz.ch>,
2026 Gary Ng <71564.1743@CompuServe.COM>,
2027 Tom Phoenix <rootbeer@teleport.com>,
2028 AndrE<eacute> Pirard <A.Pirard@ulg.ac.be>,
2029 Peter Prymmer <pvhp@forte.com>,
2030 Hugo van der Sanden <hv@crypt0.demon.co.uk>,
2031 Gurusamy Sarathy <gsar@activestate.com>,
2032 Paul J. Schinder <schinder@pobox.com>,
2033 Michael G Schwern <schwern@pobox.com>,
2034 Dan Sugalski <sugalskd@ous.edu>,
2035 Nathan Torkington <gnat@frii.com>.
2037 This document is maintained by Chris Nandor
2042 Version 1.47, last modified 22 March 2000