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 A common cause of unportable programs is the misuse of chop() to trim
107 You can get away with this on Unix and MacOS (they have a single
108 character end-of-line), but the same program will break under DOSish
109 perls because you're only chop()ing half the end-of-line. Instead,
110 chomp() should be used to trim newlines. The Dunce::Files module can
111 help audit your code for misuses of chop().
113 When dealing with binary files (or text files in binary mode) be sure
114 to explicitly set $/ to the appropriate value for your file format
115 before using chomp().
117 Because of the "text" mode translation, DOSish perls have limitations
118 in using C<seek> and C<tell> on a file accessed in "text" mode.
119 Stick to C<seek>-ing to locations you got from C<tell> (and no
120 others), and you are usually free to use C<seek> and C<tell> even
121 in "text" mode. Using C<seek> or C<tell> or other file operations
122 may be non-portable. If you use C<binmode> on a file, however, you
123 can usually C<seek> and C<tell> with arbitrary values in safety.
125 A common misconception in socket programming is that C<\n> eq C<\012>
126 everywhere. When using protocols such as common Internet protocols,
127 C<\012> and C<\015> are called for specifically, and the values of
128 the logical C<\n> and C<\r> (carriage return) are not reliable.
130 print SOCKET "Hi there, client!\r\n"; # WRONG
131 print SOCKET "Hi there, client!\015\012"; # RIGHT
133 However, using C<\015\012> (or C<\cM\cJ>, or C<\x0D\x0A>) can be tedious
134 and unsightly, as well as confusing to those maintaining the code. As
135 such, the Socket module supplies the Right Thing for those who want it.
137 use Socket qw(:DEFAULT :crlf);
138 print SOCKET "Hi there, client!$CRLF" # RIGHT
140 When reading from a socket, remember that the default input record
141 separator C<$/> is C<\n>, but robust socket code will recognize as
142 either C<\012> or C<\015\012> as end of line:
148 Because both CRLF and LF end in LF, the input record separator can
149 be set to LF and any CR stripped later. Better to write:
151 use Socket qw(:DEFAULT :crlf);
152 local($/) = LF; # not needed if $/ is already \012
155 s/$CR?$LF/\n/; # not sure if socket uses LF or CRLF, OK
156 # s/\015?\012/\n/; # same thing
159 This example is preferred over the previous one--even for Unix
160 platforms--because now any C<\015>'s (C<\cM>'s) are stripped out
161 (and there was much rejoicing).
163 Similarly, functions that return text data--such as a function that
164 fetches a web page--should sometimes translate newlines before
165 returning the data, if they've not yet been translated to the local
166 newline representation. A single line of code will often suffice:
168 $data =~ s/\015?\012/\n/g;
171 Some of this may be confusing. Here's a handy reference to the ASCII CR
172 and LF characters. You can print it out and stick it in your wallet.
174 LF == \012 == \x0A == \cJ == ASCII 10
175 CR == \015 == \x0D == \cM == ASCII 13
178 ---------------------------
181 \n * | LF | CRLF | CR |
182 \r * | CR | CR | LF |
183 ---------------------------
186 The Unix column assumes that you are not accessing a serial line
187 (like a tty) in canonical mode. If you are, then CR on input becomes
188 "\n", and "\n" on output becomes CRLF.
190 These are just the most common definitions of C<\n> and C<\r> in Perl.
191 There may well be others.
193 =head2 Numbers endianness and Width
195 Different CPUs store integers and floating point numbers in different
196 orders (called I<endianness>) and widths (32-bit and 64-bit being the
197 most common today). This affects your programs when they attempt to transfer
198 numbers in binary format from one CPU architecture to another,
199 usually either "live" via network connection, or by storing the
200 numbers to secondary storage such as a disk file or tape.
202 Conflicting storage orders make utter mess out of the numbers. If a
203 little-endian host (Intel, VAX) stores 0x12345678 (305419896 in
204 decimal), a big-endian host (Motorola, Sparc, PA) reads it as
205 0x78563412 (2018915346 in decimal). Alpha and MIPS can be either:
206 Digital/Compaq used/uses them in little-endian mode; SGI/Cray uses
207 them in big-endian mode. To avoid this problem in network (socket)
208 connections use the C<pack> and C<unpack> formats C<n> and C<N>, the
209 "network" orders. These are guaranteed to be portable.
211 You can explore the endianness of your platform by unpacking a
212 data structure packed in native format such as:
214 print unpack("h*", pack("s2", 1, 2)), "\n";
215 # '10002000' on e.g. Intel x86 or Alpha 21064 in little-endian mode
216 # '00100020' on e.g. Motorola 68040
218 If you need to distinguish between endian architectures you could use
219 either of the variables set like so:
221 $is_big_endian = unpack("h*", pack("s", 1)) =~ /01/;
222 $is_little_endian = unpack("h*", pack("s", 1)) =~ /^1/;
224 Differing widths can cause truncation even between platforms of equal
225 endianness. The platform of shorter width loses the upper parts of the
226 number. There is no good solution for this problem except to avoid
227 transferring or storing raw binary numbers.
229 One can circumnavigate both these problems in two ways. Either
230 transfer and store numbers always in text format, instead of raw
231 binary, or else consider using modules like Data::Dumper (included in
232 the standard distribution as of Perl 5.005) and Storable (included as
233 of perl 5.8). Keeping all data as text significantly simplifies matters.
235 The v-strings are portable only up to v2147483647 (0x7FFFFFFF), that's
236 how far EBCDIC, or more precisely UTF-EBCDIC will go.
238 =head2 Files and Filesystems
240 Most platforms these days structure files in a hierarchical fashion.
241 So, it is reasonably safe to assume that all platforms support the
242 notion of a "path" to uniquely identify a file on the system. How
243 that path is really written, though, differs considerably.
245 Although similar, file path specifications differ between Unix,
246 Windows, S<Mac OS>, OS/2, VMS, VOS, S<RISC OS>, and probably others.
247 Unix, for example, is one of the few OSes that has the elegant idea
248 of a single root directory.
250 DOS, OS/2, VMS, VOS, and Windows can work similarly to Unix with C</>
251 as path separator, or in their own idiosyncratic ways (such as having
252 several root directories and various "unrooted" device files such NIL:
255 S<Mac OS> uses C<:> as a path separator instead of C</>.
257 The filesystem may support neither hard links (C<link>) nor
258 symbolic links (C<symlink>, C<readlink>, C<lstat>).
260 The filesystem may support neither access timestamp nor change
261 timestamp (meaning that about the only portable timestamp is the
262 modification timestamp), or one second granularity of any timestamps
263 (e.g. the FAT filesystem limits the time granularity to two seconds).
265 VOS perl can emulate Unix filenames with C</> as path separator. The
266 native pathname characters greater-than, less-than, number-sign, and
267 percent-sign are always accepted.
269 S<RISC OS> perl can emulate Unix filenames with C</> as path
270 separator, or go native and use C<.> for path separator and C<:> to
271 signal filesystems and disk names.
273 Don't assume UNIX filesystem access semantics: that read, write,
274 and execute are all the permissions there are, and even if they exist,
275 that their semantics (for example what do r, w, and x mean on
276 a directory) are the UNIX ones. The various UNIX/POSIX compatibility
277 layers usually try to make interfaces like chmod() work, but sometimes
278 there simply is no good mapping.
280 If all this is intimidating, have no (well, maybe only a little)
281 fear. There are modules that can help. The File::Spec modules
282 provide methods to do the Right Thing on whatever platform happens
283 to be running the program.
285 use File::Spec::Functions;
286 chdir(updir()); # go up one directory
287 $file = catfile(curdir(), 'temp', 'file.txt');
288 # on Unix and Win32, './temp/file.txt'
289 # on Mac OS, ':temp:file.txt'
290 # on VMS, '[.temp]file.txt'
292 File::Spec is available in the standard distribution as of version
293 5.004_05. File::Spec::Functions is only in File::Spec 0.7 and later,
294 and some versions of perl come with version 0.6. If File::Spec
295 is not updated to 0.7 or later, you must use the object-oriented
296 interface from File::Spec (or upgrade File::Spec).
298 In general, production code should not have file paths hardcoded.
299 Making them user-supplied or read from a configuration file is
300 better, keeping in mind that file path syntax varies on different
303 This is especially noticeable in scripts like Makefiles and test suites,
304 which often assume C</> as a path separator for subdirectories.
306 Also of use is File::Basename from the standard distribution, which
307 splits a pathname into pieces (base filename, full path to directory,
310 Even when on a single platform (if you can call Unix a single platform),
311 remember not to count on the existence or the contents of particular
312 system-specific files or directories, like F</etc/passwd>,
313 F</etc/sendmail.conf>, F</etc/resolv.conf>, or even F</tmp/>. For
314 example, F</etc/passwd> may exist but not contain the encrypted
315 passwords, because the system is using some form of enhanced security.
316 Or it may not contain all the accounts, because the system is using NIS.
317 If code does need to rely on such a file, include a description of the
318 file and its format in the code's documentation, then make it easy for
319 the user to override the default location of the file.
321 Don't assume a text file will end with a newline. They should,
324 Do not have two files or directories of the same name with different
325 case, like F<test.pl> and F<Test.pl>, as many platforms have
326 case-insensitive (or at least case-forgiving) filenames. Also, try
327 not to have non-word characters (except for C<.>) in the names, and
328 keep them to the 8.3 convention, for maximum portability, onerous a
329 burden though this may appear.
331 Likewise, when using the AutoSplit module, try to keep your functions to
332 8.3 naming and case-insensitive conventions; or, at the least,
333 make it so the resulting files have a unique (case-insensitively)
336 Whitespace in filenames is tolerated on most systems, but not all,
337 and even on systems where it might be tolerated, some utilities
338 might become confused by such whitespace.
340 Many systems (DOS, VMS) cannot have more than one C<.> in their filenames.
342 Don't assume C<< > >> won't be the first character of a filename.
343 Always use C<< < >> explicitly to open a file for reading, or even
344 better, use the three-arg version of open, unless you want the user to
345 be able to specify a pipe open.
347 open(FILE, '<', $existing_file) or die $!;
349 If filenames might use strange characters, it is safest to open it
350 with C<sysopen> instead of C<open>. C<open> is magic and can
351 translate characters like C<< > >>, C<< < >>, and C<|>, which may
352 be the wrong thing to do. (Sometimes, though, it's the right thing.)
353 Three-arg open can also help protect against this translation in cases
354 where it is undesirable.
356 Don't use C<:> as a part of a filename since many systems use that for
357 their own semantics (MacOS Classic for separating pathname components,
358 many networking schemes and utilities for separating the nodename and
359 the pathname, and so on). For the same reasons, avoid C<@>, C<;> and
362 The I<portable filename characters> as defined by ANSI C are
364 a b c d e f g h i j k l m n o p q r t u v w x y z
365 A B C D E F G H I J K L M N O P Q R T U V W X Y Z
369 and the "-" shouldn't be the first character. If you want to be
370 hypercorrect, stay within the 8.3 naming convention (all the files and
371 directories have to be unique within one directory if their names are
372 lowercased and truncated to eight characters before the C<.>, if any,
373 and to three characters after the C<.>, if any). (And do not use
374 C<.>s in directory names.)
376 =head2 System Interaction
378 Not all platforms provide a command line. These are usually platforms
379 that rely primarily on a Graphical User Interface (GUI) for user
380 interaction. A program requiring a command line interface might
381 not work everywhere. This is probably for the user of the program
382 to deal with, so don't stay up late worrying about it.
384 Some platforms can't delete or rename files held open by the system.
385 Remember to C<close> files when you are done with them. Don't
386 C<unlink> or C<rename> an open file. Don't C<tie> or C<open> a
387 file already tied or opened; C<untie> or C<close> it first.
389 Don't open the same file more than once at a time for writing, as some
390 operating systems put mandatory locks on such files.
392 Don't assume that write/modify permission on a directory gives the
393 right to add or delete files/directories in that directory. That is
394 filesystem specific: in some filesystems you need write/modify
395 permission also (or even just) in the file/directory itself. In some
396 filesystems (AFS, DFS) the permission to add/delete directory entries
397 is a completely separate permission.
399 Don't assume that a single C<unlink> completely gets rid of the file:
400 some filesystems (most notably the ones in VMS) have versioned
401 filesystems, and unlink() removes only the most recent one (it doesn't
402 remove all the versions because by default the native tools on those
403 platforms remove just the most recent version, too). The portable
404 idiom to remove all the versions of a file is
406 1 while unlink "file";
408 This will terminate if the file is undeleteable for some reason
409 (protected, not there, and so on).
411 Don't count on a specific environment variable existing in C<%ENV>.
412 Don't count on C<%ENV> entries being case-sensitive, or even
413 case-preserving. Don't try to clear %ENV by saying C<%ENV = ();>, or,
414 if you really have to, make it conditional on C<$^O ne 'VMS'> since in
415 VMS the C<%ENV> table is much more than a per-process key-value string
418 Don't count on signals or C<%SIG> for anything.
420 Don't count on filename globbing. Use C<opendir>, C<readdir>, and
423 Don't count on per-program environment variables, or per-program current
426 Don't count on specific values of C<$!>.
428 =head2 Interprocess Communication (IPC)
430 In general, don't directly access the system in code meant to be
431 portable. That means, no C<system>, C<exec>, C<fork>, C<pipe>,
432 C<``>, C<qx//>, C<open> with a C<|>, nor any of the other things
433 that makes being a perl hacker worth being.
435 Commands that launch external processes are generally supported on
436 most platforms (though many of them do not support any type of
437 forking). The problem with using them arises from what you invoke
438 them on. External tools are often named differently on different
439 platforms, may not be available in the same location, might accept
440 different arguments, can behave differently, and often present their
441 results in a platform-dependent way. Thus, you should seldom depend
442 on them to produce consistent results. (Then again, if you're calling
443 I<netstat -a>, you probably don't expect it to run on both Unix and CP/M.)
445 One especially common bit of Perl code is opening a pipe to B<sendmail>:
447 open(MAIL, '|/usr/lib/sendmail -t')
448 or die "cannot fork sendmail: $!";
450 This is fine for systems programming when sendmail is known to be
451 available. But it is not fine for many non-Unix systems, and even
452 some Unix systems that may not have sendmail installed. If a portable
453 solution is needed, see the various distributions on CPAN that deal
454 with it. Mail::Mailer and Mail::Send in the MailTools distribution are
455 commonly used, and provide several mailing methods, including mail,
456 sendmail, and direct SMTP (via Net::SMTP) if a mail transfer agent is
457 not available. Mail::Sendmail is a standalone module that provides
458 simple, platform-independent mailing.
460 The Unix System V IPC (C<msg*(), sem*(), shm*()>) is not available
461 even on all Unix platforms.
463 Do not use either the bare result of C<pack("N", 10, 20, 30, 40)> or
464 bare v-strings (such as C<v10.20.30.40>) to represent IPv4 addresses:
465 both forms just pack the four bytes into network order. That this
466 would be equal to the C language C<in_addr> struct (which is what the
467 socket code internally uses) is not guaranteed. To be portable use
468 the routines of the Socket extension, such as C<inet_aton()>,
469 C<inet_ntoa()>, and C<sockaddr_in()>.
471 The rule of thumb for portable code is: Do it all in portable Perl, or
472 use a module (that may internally implement it with platform-specific
473 code, but expose a common interface).
475 =head2 External Subroutines (XS)
477 XS code can usually be made to work with any platform, but dependent
478 libraries, header files, etc., might not be readily available or
479 portable, or the XS code itself might be platform-specific, just as Perl
480 code might be. If the libraries and headers are portable, then it is
481 normally reasonable to make sure the XS code is portable, too.
483 A different type of portability issue arises when writing XS code:
484 availability of a C compiler on the end-user's system. C brings
485 with it its own portability issues, and writing XS code will expose
486 you to some of those. Writing purely in Perl is an easier way to
489 =head2 Standard Modules
491 In general, the standard modules work across platforms. Notable
492 exceptions are the CPAN module (which currently makes connections to external
493 programs that may not be available), platform-specific modules (like
494 ExtUtils::MM_VMS), and DBM modules.
496 There is no one DBM module available on all platforms.
497 SDBM_File and the others are generally available on all Unix and DOSish
498 ports, but not in MacPerl, where only NBDM_File and DB_File are
501 The good news is that at least some DBM module should be available, and
502 AnyDBM_File will use whichever module it can find. Of course, then
503 the code needs to be fairly strict, dropping to the greatest common
504 factor (e.g., not exceeding 1K for each record), so that it will
505 work with any DBM module. See L<AnyDBM_File> for more details.
509 The system's notion of time of day and calendar date is controlled in
510 widely different ways. Don't assume the timezone is stored in C<$ENV{TZ}>,
511 and even if it is, don't assume that you can control the timezone through
514 Don't assume that the epoch starts at 00:00:00, January 1, 1970,
515 because that is OS- and implementation-specific. It is better to store a date
516 in an unambiguous representation. The ISO-8601 standard defines
517 "YYYY-MM-DD" as the date format. A text representation (like "1987-12-18")
518 can be easily converted into an OS-specific value using a module like
519 Date::Parse. An array of values, such as those returned by
520 C<localtime>, can be converted to an OS-specific representation using
523 When calculating specific times, such as for tests in time or date modules,
524 it may be appropriate to calculate an offset for the epoch.
527 $offset = Time::Local::timegm(0, 0, 0, 1, 0, 70);
529 The value for C<$offset> in Unix will be C<0>, but in Mac OS will be
530 some large number. C<$offset> can then be added to a Unix time value
531 to get what should be the proper value on any system.
533 =head2 Character sets and character encoding
535 Assume very little about character sets.
537 Assume nothing about numerical values (C<ord>, C<chr>) of characters.
538 Do not use explicit code point ranges (like \xHH-\xHH); use for
539 example symbolic character classes like C<[:print:]>.
541 Do not assume that the alphabetic characters are encoded contiguously
542 (in the numeric sense). There may be gaps.
544 Do not assume anything about the ordering of the characters.
545 The lowercase letters may come before or after the uppercase letters;
546 the lowercase and uppercase may be interlaced so that both `a' and `A'
547 come before `b'; the accented and other international characters may
548 be interlaced so that E<auml> comes before `b'.
550 =head2 Internationalisation
552 If you may assume POSIX (a rather large assumption), you may read
553 more about the POSIX locale system from L<perllocale>. The locale
554 system at least attempts to make things a little bit more portable,
555 or at least more convenient and native-friendly for non-English
556 users. The system affects character sets and encoding, and date
557 and time formatting--amongst other things.
559 =head2 System Resources
561 If your code is destined for systems with severely constrained (or
562 missing!) virtual memory systems then you want to be I<especially> mindful
563 of avoiding wasteful constructs such as:
565 # NOTE: this is no longer "bad" in perl5.005
566 for (0..10000000) {} # bad
567 for (my $x = 0; $x <= 10000000; ++$x) {} # good
569 @lines = <VERY_LARGE_FILE>; # bad
571 while (<FILE>) {$file .= $_} # sometimes bad
572 $file = join('', <FILE>); # better
574 The last two constructs may appear unintuitive to most people. The
575 first repeatedly grows a string, whereas the second allocates a
576 large chunk of memory in one go. On some systems, the second is
577 more efficient that the first.
581 Most multi-user platforms provide basic levels of security, usually
582 implemented at the filesystem level. Some, however, do
583 not-- unfortunately. Thus the notion of user id, or "home" directory,
584 or even the state of being logged-in, may be unrecognizable on many
585 platforms. If you write programs that are security-conscious, it
586 is usually best to know what type of system you will be running
587 under so that you can write code explicitly for that platform (or
590 Don't assume the UNIX filesystem access semantics: the operating
591 system or the filesystem may be using some ACL systems, which are
592 richer languages than the usual rwx. Even if the rwx exist,
593 their semantics might be different.
595 (From security viewpoint testing for permissions before attempting to
596 do something is silly anyway: if one tries this, there is potential
597 for race conditions-- someone or something might change the
598 permissions between the permissions check and the actual operation.
599 Just try the operation.)
601 Don't assume the UNIX user and group semantics: especially, don't
602 expect the C<< $< >> and C<< $> >> (or the C<$(> and C<$)>) to work
603 for switching identities (or memberships).
605 Don't assume set-uid and set-gid semantics. (And even if you do,
606 think twice: set-uid and set-gid are a known can of security worms.)
610 For those times when it is necessary to have platform-specific code,
611 consider keeping the platform-specific code in one place, making porting
612 to other platforms easier. Use the Config module and the special
613 variable C<$^O> to differentiate platforms, as described in
616 Be careful in the tests you supply with your module or programs.
617 Module code may be fully portable, but its tests might not be. This
618 often happens when tests spawn off other processes or call external
619 programs to aid in the testing, or when (as noted above) the tests
620 assume certain things about the filesystem and paths. Be careful
621 not to depend on a specific output style for errors, such as when
622 checking C<$!> after a system call. Some platforms expect a certain
623 output format, and perl on those platforms may have been adjusted
624 accordingly. Most specifically, don't anchor a regex when testing
629 Modules uploaded to CPAN are tested by a variety of volunteers on
630 different platforms. These CPAN testers are notified by mail of each
631 new upload, and reply to the list with PASS, FAIL, NA (not applicable to
632 this platform), or UNKNOWN (unknown), along with any relevant notations.
634 The purpose of the testing is twofold: one, to help developers fix any
635 problems in their code that crop up because of lack of testing on other
636 platforms; two, to provide users with information about whether
637 a given module works on a given platform.
641 =item Mailing list: cpan-testers@perl.org
643 =item Testing results: http://testers.cpan.org/
649 As of version 5.002, Perl is built with a C<$^O> variable that
650 indicates the operating system it was built on. This was implemented
651 to help speed up code that would otherwise have to C<use Config>
652 and use the value of C<$Config{osname}>. Of course, to get more
653 detailed information about the system, looking into C<%Config> is
654 certainly recommended.
656 C<%Config> cannot always be trusted, however, because it was built
657 at compile time. If perl was built in one place, then transferred
658 elsewhere, some values may be wrong. The values may even have been
659 edited after the fact.
663 Perl works on a bewildering variety of Unix and Unix-like platforms (see
664 e.g. most of the files in the F<hints/> directory in the source code kit).
665 On most of these systems, the value of C<$^O> (hence C<$Config{'osname'}>,
666 too) is determined either by lowercasing and stripping punctuation from the
667 first field of the string returned by typing C<uname -a> (or a similar command)
668 at the shell prompt or by testing the file system for the presence of
669 uniquely named files such as a kernel or header file. Here, for example,
670 are a few of the more popular Unix flavors:
672 uname $^O $Config{'archname'}
673 --------------------------------------------
675 BSD/OS bsdos i386-bsdos
676 dgux dgux AViiON-dgux
677 DYNIX/ptx dynixptx i386-dynixptx
678 FreeBSD freebsd freebsd-i386
679 Linux linux arm-linux
680 Linux linux i386-linux
681 Linux linux i586-linux
682 Linux linux ppc-linux
683 HP-UX hpux PA-RISC1.1
685 Mac OS X darwin darwin
686 MachTen PPC machten powerpc-machten
688 NeXT 4 next OPENSTEP-Mach
689 openbsd openbsd i386-openbsd
690 OSF1 dec_osf alpha-dec_osf
691 reliantunix-n svr4 RM400-svr4
692 SCO_SV sco_sv i386-sco_sv
693 SINIX-N svr4 RM400-svr4
694 sn4609 unicos CRAY_C90-unicos
695 sn6521 unicosmk t3e-unicosmk
696 sn9617 unicos CRAY_J90-unicos
697 SunOS solaris sun4-solaris
698 SunOS solaris i86pc-solaris
699 SunOS4 sunos sun4-sunos
701 Because the value of C<$Config{archname}> may depend on the
702 hardware architecture, it can vary more than the value of C<$^O>.
704 =head2 DOS and Derivatives
706 Perl has long been ported to Intel-style microcomputers running under
707 systems like PC-DOS, MS-DOS, OS/2, and most Windows platforms you can
708 bring yourself to mention (except for Windows CE, if you count that).
709 Users familiar with I<COMMAND.COM> or I<CMD.EXE> style shells should
710 be aware that each of these file specifications may have subtle
713 $filespec0 = "c:/foo/bar/file.txt";
714 $filespec1 = "c:\\foo\\bar\\file.txt";
715 $filespec2 = 'c:\foo\bar\file.txt';
716 $filespec3 = 'c:\\foo\\bar\\file.txt';
718 System calls accept either C</> or C<\> as the path separator.
719 However, many command-line utilities of DOS vintage treat C</> as
720 the option prefix, so may get confused by filenames containing C</>.
721 Aside from calling any external programs, C</> will work just fine,
722 and probably better, as it is more consistent with popular usage,
723 and avoids the problem of remembering what to backwhack and what
726 The DOS FAT filesystem can accommodate only "8.3" style filenames. Under
727 the "case-insensitive, but case-preserving" HPFS (OS/2) and NTFS (NT)
728 filesystems you may have to be careful about case returned with functions
729 like C<readdir> or used with functions like C<open> or C<opendir>.
731 DOS also treats several filenames as special, such as AUX, PRN,
732 NUL, CON, COM1, LPT1, LPT2, etc. Unfortunately, sometimes these
733 filenames won't even work if you include an explicit directory
734 prefix. It is best to avoid such filenames, if you want your code
735 to be portable to DOS and its derivatives. It's hard to know what
736 these all are, unfortunately.
738 Users of these operating systems may also wish to make use of
739 scripts such as I<pl2bat.bat> or I<pl2cmd> to
740 put wrappers around your scripts.
742 Newline (C<\n>) is translated as C<\015\012> by STDIO when reading from
743 and writing to files (see L<"Newlines">). C<binmode(FILEHANDLE)>
744 will keep C<\n> translated as C<\012> for that filehandle. Since it is a
745 no-op on other systems, C<binmode> should be used for cross-platform code
746 that deals with binary data. That's assuming you realize in advance
747 that your data is in binary. General-purpose programs should
748 often assume nothing about their data.
750 The C<$^O> variable and the C<$Config{archname}> values for various
751 DOSish perls are as follows:
753 OS $^O $Config{archname} ID Version
754 --------------------------------------------------------
758 Windows 3.1 ? ? 0 3 01
759 Windows 95 MSWin32 MSWin32-x86 1 4 00
760 Windows 98 MSWin32 MSWin32-x86 1 4 10
761 Windows ME MSWin32 MSWin32-x86 1 ?
762 Windows NT MSWin32 MSWin32-x86 2 4 xx
763 Windows NT MSWin32 MSWin32-ALPHA 2 4 xx
764 Windows NT MSWin32 MSWin32-ppc 2 4 xx
765 Windows 2000 MSWin32 MSWin32-x86 2 5 xx
766 Windows XP MSWin32 MSWin32-x86 2 ?
767 Windows CE MSWin32 ? 3
770 The various MSWin32 Perl's can distinguish the OS they are running on
771 via the value of the fifth element of the list returned from
772 Win32::GetOSVersion(). For example:
774 if ($^O eq 'MSWin32') {
775 my @os_version_info = Win32::GetOSVersion();
776 print +('3.1','95','NT')[$os_version_info[4]],"\n";
785 The djgpp environment for DOS, http://www.delorie.com/djgpp/
790 The EMX environment for DOS, OS/2, etc. emx@iaehv.nl,
791 http://www.leo.org/pub/comp/os/os2/leo/gnu/emx+gcc/index.html or
792 ftp://hobbes.nmsu.edu/pub/os2/dev/emx. Also L<perlos2>.
796 Build instructions for Win32 in L<perlwin32>, or under the Cygnus environment
801 The C<Win32::*> modules in L<Win32>.
805 The ActiveState Pages, http://www.activestate.com/
809 The Cygwin environment for Win32; F<README.cygwin> (installed
810 as L<perlcygwin>), http://www.cygwin.com/
814 The U/WIN environment for Win32,
815 http://www.research.att.com/sw/tools/uwin/
819 Build instructions for OS/2, L<perlos2>
825 Any module requiring XS compilation is right out for most people, because
826 MacPerl is built using non-free (and non-cheap!) compilers. Some XS
827 modules that can work with MacPerl are built and distributed in binary
830 Directories are specified as:
832 volume:folder:file for absolute pathnames
833 volume:folder: for absolute pathnames
834 :folder:file for relative pathnames
835 :folder: for relative pathnames
836 :file for relative pathnames
837 file for relative pathnames
839 Files are stored in the directory in alphabetical order. Filenames are
840 limited to 31 characters, and may include any character except for
841 null and C<:>, which is reserved as the path separator.
843 Instead of C<flock>, see C<FSpSetFLock> and C<FSpRstFLock> in the
844 Mac::Files module, or C<chmod(0444, ...)> and C<chmod(0666, ...)>.
846 In the MacPerl application, you can't run a program from the command line;
847 programs that expect C<@ARGV> to be populated can be edited with something
848 like the following, which brings up a dialog box asking for the command
852 @ARGV = split /\s+/, MacPerl::Ask('Arguments?');
855 A MacPerl script saved as a "droplet" will populate C<@ARGV> with the full
856 pathnames of the files dropped onto the script.
858 Mac users can run programs under a type of command line interface
859 under MPW (Macintosh Programmer's Workshop, a free development
860 environment from Apple). MacPerl was first introduced as an MPW
861 tool, and MPW can be used like a shell:
863 perl myscript.plx some arguments
865 ToolServer is another app from Apple that provides access to MPW tools
866 from MPW and the MacPerl app, which allows MacPerl programs to use
867 C<system>, backticks, and piped C<open>.
869 "S<Mac OS>" is the proper name for the operating system, but the value
870 in C<$^O> is "MacOS". To determine architecture, version, or whether
871 the application or MPW tool version is running, check:
873 $is_app = $MacPerl::Version =~ /App/;
874 $is_tool = $MacPerl::Version =~ /MPW/;
875 ($version) = $MacPerl::Version =~ /^(\S+)/;
876 $is_ppc = $MacPerl::Architecture eq 'MacPPC';
877 $is_68k = $MacPerl::Architecture eq 'Mac68K';
879 S<Mac OS X>, based on NeXT's OpenStep OS, runs MacPerl natively, under the
880 "Classic" environment. There is no "Carbon" version of MacPerl to run
881 under the primary Mac OS X environment. S<Mac OS X> and its Open Source
882 version, Darwin, both run Unix perl natively.
890 MacPerl Development, http://dev.macperl.org/ .
894 The MacPerl Pages, http://www.macperl.com/ .
898 The MacPerl mailing lists, http://lists.perl.org/ .
904 Perl on VMS is discussed in L<perlvms> in the perl distribution.
905 Perl on VMS can accept either VMS- or Unix-style file
906 specifications as in either of the following:
908 $ perl -ne "print if /perl_setup/i" SYS$LOGIN:LOGIN.COM
909 $ perl -ne "print if /perl_setup/i" /sys$login/login.com
911 but not a mixture of both as in:
913 $ perl -ne "print if /perl_setup/i" sys$login:/login.com
914 Can't open sys$login:/login.com: file specification syntax error
916 Interacting with Perl from the Digital Command Language (DCL) shell
917 often requires a different set of quotation marks than Unix shells do.
920 $ perl -e "print ""Hello, world.\n"""
923 There are several ways to wrap your perl scripts in DCL F<.COM> files, if
924 you are so inclined. For example:
926 $ write sys$output "Hello from DCL!"
928 $ then perl -x 'f$environment("PROCEDURE")
929 $ else perl -x - 'p1 'p2 'p3 'p4 'p5 'p6 'p7 'p8
930 $ deck/dollars="__END__"
933 print "Hello from Perl!\n";
938 Do take care with C<$ ASSIGN/nolog/user SYS$COMMAND: SYS$INPUT> if your
939 perl-in-DCL script expects to do things like C<< $read = <STDIN>; >>.
941 Filenames are in the format "name.extension;version". The maximum
942 length for filenames is 39 characters, and the maximum length for
943 extensions is also 39 characters. Version is a number from 1 to
944 32767. Valid characters are C</[A-Z0-9$_-]/>.
946 VMS's RMS filesystem is case-insensitive and does not preserve case.
947 C<readdir> returns lowercased filenames, but specifying a file for
948 opening remains case-insensitive. Files without extensions have a
949 trailing period on them, so doing a C<readdir> with a file named F<A.;5>
950 will return F<a.> (though that file could be opened with
953 RMS had an eight level limit on directory depths from any rooted logical
954 (allowing 16 levels overall) prior to VMS 7.2. Hence
955 C<PERL_ROOT:[LIB.2.3.4.5.6.7.8]> is a valid directory specification but
956 C<PERL_ROOT:[LIB.2.3.4.5.6.7.8.9]> is not. F<Makefile.PL> authors might
957 have to take this into account, but at least they can refer to the former
958 as C</PERL_ROOT/lib/2/3/4/5/6/7/8/>.
960 The VMS::Filespec module, which gets installed as part of the build
961 process on VMS, is a pure Perl module that can easily be installed on
962 non-VMS platforms and can be helpful for conversions to and from RMS
965 What C<\n> represents depends on the type of file opened. It usually
966 represents C<\012> but it could also be C<\015>, C<\012>, C<\015\012>,
967 C<\000>, C<\040>, or nothing depending on the file organiztion and
968 record format. The VMS::Stdio module provides access to the
969 special fopen() requirements of files with unusual attributes on VMS.
971 TCP/IP stacks are optional on VMS, so socket routines might not be
972 implemented. UDP sockets may not be supported.
974 The value of C<$^O> on OpenVMS is "VMS". To determine the architecture
975 that you are running on without resorting to loading all of C<%Config>
976 you can examine the content of the C<@INC> array like so:
978 if (grep(/VMS_AXP/, @INC)) {
979 print "I'm on Alpha!\n";
981 } elsif (grep(/VMS_VAX/, @INC)) {
982 print "I'm on VAX!\n";
985 print "I'm not so sure about where $^O is...\n";
988 On VMS, perl determines the UTC offset from the C<SYS$TIMEZONE_DIFFERENTIAL>
989 logical name. Although the VMS epoch began at 17-NOV-1858 00:00:00.00,
990 calls to C<localtime> are adjusted to count offsets from
991 01-JAN-1970 00:00:00.00, just like Unix.
999 F<README.vms> (installed as L<README_vms>), L<perlvms>
1003 vmsperl list, majordomo@perl.org
1005 (Put the words C<subscribe vmsperl> in message body.)
1009 vmsperl on the web, http://www.sidhe.org/vmsperl/index.html
1015 Perl on VOS is discussed in F<README.vos> in the perl distribution
1016 (installed as L<perlvos>). Perl on VOS can accept either VOS- or
1017 Unix-style file specifications as in either of the following:
1019 $ perl -ne "print if /perl_setup/i" >system>notices
1020 $ perl -ne "print if /perl_setup/i" /system/notices
1022 or even a mixture of both as in:
1024 $ perl -ne "print if /perl_setup/i" >system/notices
1026 Even though VOS allows the slash character to appear in object
1027 names, because the VOS port of Perl interprets it as a pathname
1028 delimiting character, VOS files, directories, or links whose names
1029 contain a slash character cannot be processed. Such files must be
1030 renamed before they can be processed by Perl. Note that VOS limits
1031 file names to 32 or fewer characters.
1033 See F<README.vos> for restrictions that apply when Perl is built
1034 with the alpha version of VOS POSIX.1 support.
1036 Perl on VOS is built without any extensions and does not support
1039 The value of C<$^O> on VOS is "VOS". To determine the architecture that
1040 you are running on without resorting to loading all of C<%Config> you
1041 can examine the content of the @INC array like so:
1044 print "I'm on a Stratus box!\n";
1046 print "I'm not on a Stratus box!\n";
1050 if (grep(/860/, @INC)) {
1051 print "This box is a Stratus XA/R!\n";
1053 } elsif (grep(/7100/, @INC)) {
1054 print "This box is a Stratus HP 7100 or 8xxx!\n";
1056 } elsif (grep(/8000/, @INC)) {
1057 print "This box is a Stratus HP 8xxx!\n";
1060 print "This box is a Stratus 68K!\n";
1073 The VOS mailing list.
1075 There is no specific mailing list for Perl on VOS. You can post
1076 comments to the comp.sys.stratus newsgroup, or subscribe to the general
1077 Stratus mailing list. Send a letter with "Subscribe Info-Stratus" in
1078 the message body to majordomo@list.stratagy.com.
1082 VOS Perl on the web at http://ftp.stratus.com/pub/vos/vos.html
1086 =head2 EBCDIC Platforms
1088 Recent versions of Perl have been ported to platforms such as OS/400 on
1089 AS/400 minicomputers as well as OS/390, VM/ESA, and BS2000 for S/390
1090 Mainframes. Such computers use EBCDIC character sets internally (usually
1091 Character Code Set ID 0037 for OS/400 and either 1047 or POSIX-BC for S/390
1092 systems). On the mainframe perl currently works under the "Unix system
1093 services for OS/390" (formerly known as OpenEdition), VM/ESA OpenEdition, or
1094 the BS200 POSIX-BC system (BS2000 is supported in perl 5.6 and greater).
1095 See L<perlos390> for details.
1097 As of R2.5 of USS for OS/390 and Version 2.3 of VM/ESA these Unix
1098 sub-systems do not support the C<#!> shebang trick for script invocation.
1099 Hence, on OS/390 and VM/ESA perl scripts can be executed with a header
1100 similar to the following simple script:
1103 eval 'exec /usr/local/bin/perl -S $0 ${1+"$@"}'
1105 #!/usr/local/bin/perl # just a comment really
1107 print "Hello from perl!\n";
1109 OS/390 will support the C<#!> shebang trick in release 2.8 and beyond.
1110 Calls to C<system> and backticks can use POSIX shell syntax on all
1113 On the AS/400, if PERL5 is in your library list, you may need
1114 to wrap your perl scripts in a CL procedure to invoke them like so:
1117 CALL PGM(PERL5/PERL) PARM('/QOpenSys/hello.pl')
1120 This will invoke the perl script F<hello.pl> in the root of the
1121 QOpenSys file system. On the AS/400 calls to C<system> or backticks
1124 On these platforms, bear in mind that the EBCDIC character set may have
1125 an effect on what happens with some perl functions (such as C<chr>,
1126 C<pack>, C<print>, C<printf>, C<ord>, C<sort>, C<sprintf>, C<unpack>), as
1127 well as bit-fiddling with ASCII constants using operators like C<^>, C<&>
1128 and C<|>, not to mention dealing with socket interfaces to ASCII computers
1129 (see L<"Newlines">).
1131 Fortunately, most web servers for the mainframe will correctly
1132 translate the C<\n> in the following statement to its ASCII equivalent
1133 (C<\r> is the same under both Unix and OS/390 & VM/ESA):
1135 print "Content-type: text/html\r\n\r\n";
1137 The values of C<$^O> on some of these platforms includes:
1139 uname $^O $Config{'archname'}
1140 --------------------------------------------
1143 POSIX-BC posix-bc BS2000-posix-bc
1146 Some simple tricks for determining if you are running on an EBCDIC
1147 platform could include any of the following (perhaps all):
1149 if ("\t" eq "\05") { print "EBCDIC may be spoken here!\n"; }
1151 if (ord('A') == 193) { print "EBCDIC may be spoken here!\n"; }
1153 if (chr(169) eq 'z') { print "EBCDIC may be spoken here!\n"; }
1155 One thing you may not want to rely on is the EBCDIC encoding
1156 of punctuation characters since these may differ from code page to code
1157 page (and once your module or script is rumoured to work with EBCDIC,
1158 folks will want it to work with all EBCDIC character sets).
1168 L<perlos390>, F<README.os390>, F<perlbs2000>, F<README.vmesa>,
1173 The perl-mvs@perl.org list is for discussion of porting issues as well as
1174 general usage issues for all EBCDIC Perls. Send a message body of
1175 "subscribe perl-mvs" to majordomo@perl.org.
1179 AS/400 Perl information at
1180 http://as400.rochester.ibm.com/
1181 as well as on CPAN in the F<ports/> directory.
1185 =head2 Acorn RISC OS
1187 Because Acorns use ASCII with newlines (C<\n>) in text files as C<\012> like
1188 Unix, and because Unix filename emulation is turned on by default,
1189 most simple scripts will probably work "out of the box". The native
1190 filesystem is modular, and individual filesystems are free to be
1191 case-sensitive or insensitive, and are usually case-preserving. Some
1192 native filesystems have name length limits, which file and directory
1193 names are silently truncated to fit. Scripts should be aware that the
1194 standard filesystem currently has a name length limit of B<10>
1195 characters, with up to 77 items in a directory, but other filesystems
1196 may not impose such limitations.
1198 Native filenames are of the form
1200 Filesystem#Special_Field::DiskName.$.Directory.Directory.File
1204 Special_Field is not usually present, but may contain . and $ .
1205 Filesystem =~ m|[A-Za-z0-9_]|
1206 DsicName =~ m|[A-Za-z0-9_/]|
1207 $ represents the root directory
1208 . is the path separator
1209 @ is the current directory (per filesystem but machine global)
1210 ^ is the parent directory
1211 Directory and File =~ m|[^\0- "\.\$\%\&:\@\\^\|\177]+|
1213 The default filename translation is roughly C<tr|/.|./|;>
1215 Note that C<"ADFS::HardDisk.$.File" ne 'ADFS::HardDisk.$.File'> and that
1216 the second stage of C<$> interpolation in regular expressions will fall
1217 foul of the C<$.> if scripts are not careful.
1219 Logical paths specified by system variables containing comma-separated
1220 search lists are also allowed; hence C<System:Modules> is a valid
1221 filename, and the filesystem will prefix C<Modules> with each section of
1222 C<System$Path> until a name is made that points to an object on disk.
1223 Writing to a new file C<System:Modules> would be allowed only if
1224 C<System$Path> contains a single item list. The filesystem will also
1225 expand system variables in filenames if enclosed in angle brackets, so
1226 C<< <System$Dir>.Modules >> would look for the file
1227 S<C<$ENV{'System$Dir'} . 'Modules'>>. The obvious implication of this is
1228 that B<fully qualified filenames can start with C<< <> >>> and should
1229 be protected when C<open> is used for input.
1231 Because C<.> was in use as a directory separator and filenames could not
1232 be assumed to be unique after 10 characters, Acorn implemented the C
1233 compiler to strip the trailing C<.c> C<.h> C<.s> and C<.o> suffix from
1234 filenames specified in source code and store the respective files in
1235 subdirectories named after the suffix. Hence files are translated:
1238 C:foo.h C:h.foo (logical path variable)
1239 sys/os.h sys.h.os (C compiler groks Unix-speak)
1240 10charname.c c.10charname
1241 10charname.o o.10charname
1242 11charname_.c c.11charname (assuming filesystem truncates at 10)
1244 The Unix emulation library's translation of filenames to native assumes
1245 that this sort of translation is required, and it allows a user-defined list
1246 of known suffixes that it will transpose in this fashion. This may
1247 seem transparent, but consider that with these rules C<foo/bar/baz.h>
1248 and C<foo/bar/h/baz> both map to C<foo.bar.h.baz>, and that C<readdir> and
1249 C<glob> cannot and do not attempt to emulate the reverse mapping. Other
1250 C<.>'s in filenames are translated to C</>.
1252 As implied above, the environment accessed through C<%ENV> is global, and
1253 the convention is that program specific environment variables are of the
1254 form C<Program$Name>. Each filesystem maintains a current directory,
1255 and the current filesystem's current directory is the B<global> current
1256 directory. Consequently, sociable programs don't change the current
1257 directory but rely on full pathnames, and programs (and Makefiles) cannot
1258 assume that they can spawn a child process which can change the current
1259 directory without affecting its parent (and everyone else for that
1262 Because native operating system filehandles are global and are currently
1263 allocated down from 255, with 0 being a reserved value, the Unix emulation
1264 library emulates Unix filehandles. Consequently, you can't rely on
1265 passing C<STDIN>, C<STDOUT>, or C<STDERR> to your children.
1267 The desire of users to express filenames of the form
1268 C<< <Foo$Dir>.Bar >> on the command line unquoted causes problems,
1269 too: C<``> command output capture has to perform a guessing game. It
1270 assumes that a string C<< <[^<>]+\$[^<>]> >> is a
1271 reference to an environment variable, whereas anything else involving
1272 C<< < >> or C<< > >> is redirection, and generally manages to be 99%
1273 right. Of course, the problem remains that scripts cannot rely on any
1274 Unix tools being available, or that any tools found have Unix-like command
1277 Extensions and XS are, in theory, buildable by anyone using free
1278 tools. In practice, many don't, as users of the Acorn platform are
1279 used to binary distributions. MakeMaker does run, but no available
1280 make currently copes with MakeMaker's makefiles; even if and when
1281 this should be fixed, the lack of a Unix-like shell will cause
1282 problems with makefile rules, especially lines of the form C<cd
1283 sdbm && make all>, and anything using quoting.
1285 "S<RISC OS>" is the proper name for the operating system, but the value
1286 in C<$^O> is "riscos" (because we don't like shouting).
1290 Perl has been ported to many platforms that do not fit into any of
1291 the categories listed above. Some, such as AmigaOS, Atari MiNT,
1292 BeOS, HP MPE/iX, QNX, Plan 9, and VOS, have been well-integrated
1293 into the standard Perl source code kit. You may need to see the
1294 F<ports/> directory on CPAN for information, and possibly binaries,
1295 for the likes of: aos, Atari ST, lynxos, riscos, Novell Netware,
1296 Tandem Guardian, I<etc.> (Yes, we know that some of these OSes may
1297 fall under the Unix category, but we are not a standards body.)
1299 Some approximate operating system names and their C<$^O> values
1300 in the "OTHER" category include:
1302 OS $^O $Config{'archname'}
1303 ------------------------------------------
1304 Amiga DOS amigaos m68k-amigos
1305 MPE/iX mpeix PA-RISC1.1
1313 Amiga, F<README.amiga> (installed as L<perlamiga>).
1317 Atari, F<README.mint> and Guido Flohr's web page
1318 http://stud.uni-sb.de/~gufl0000/
1322 Be OS, F<README.beos>
1326 HP 300 MPE/iX, F<README.mpeix> and Mark Bixby's web page
1327 http://www.bixby.org/mark/perlix.html
1331 A free perl5-based PERL.NLM for Novell Netware is available in
1332 precompiled binary and source code form from http://www.novell.com/
1333 as well as from CPAN.
1337 Plan 9, F<README.plan9>
1341 =head1 FUNCTION IMPLEMENTATIONS
1343 Listed below are functions that are either completely unimplemented
1344 or else have been implemented differently on various platforms.
1345 Following each description will be, in parentheses, a list of
1346 platforms that the description applies to.
1348 The list may well be incomplete, or even wrong in some places. When
1349 in doubt, consult the platform-specific README files in the Perl
1350 source distribution, and any other documentation resources accompanying
1353 Be aware, moreover, that even among Unix-ish systems there are variations.
1355 For many functions, you can also query C<%Config>, exported by
1356 default from the Config module. For example, to check whether the
1357 platform has the C<lstat> call, check C<$Config{d_lstat}>. See
1358 L<Config> for a full description of available variables.
1360 =head2 Alphabetical Listing of Perl Functions
1370 C<-r>, C<-w>, and C<-x> have a limited meaning only; directories
1371 and applications are executable, and there are no uid/gid
1372 considerations. C<-o> is not supported. (S<Mac OS>)
1374 C<-r>, C<-w>, C<-x>, and C<-o> tell whether the file is accessible,
1375 which may not reflect UIC-based file protections. (VMS)
1377 C<-s> returns the size of the data fork, not the total size of data fork
1378 plus resource fork. (S<Mac OS>).
1380 C<-s> by name on an open file will return the space reserved on disk,
1381 rather than the current extent. C<-s> on an open filehandle returns the
1382 current size. (S<RISC OS>)
1384 C<-R>, C<-W>, C<-X>, C<-O> are indistinguishable from C<-r>, C<-w>,
1385 C<-x>, C<-o>. (S<Mac OS>, Win32, VMS, S<RISC OS>)
1387 C<-b>, C<-c>, C<-k>, C<-g>, C<-p>, C<-u>, C<-A> are not implemented.
1390 C<-g>, C<-k>, C<-l>, C<-p>, C<-u>, C<-A> are not particularly meaningful.
1391 (Win32, VMS, S<RISC OS>)
1393 C<-d> is true if passed a device spec without an explicit directory.
1396 C<-T> and C<-B> are implemented, but might misclassify Mac text files
1397 with foreign characters; this is the case will all platforms, but may
1398 affect S<Mac OS> often. (S<Mac OS>)
1400 C<-x> (or C<-X>) determine if a file ends in one of the executable
1401 suffixes. C<-S> is meaningless. (Win32)
1403 C<-x> (or C<-X>) determine if a file has an executable file type.
1410 Not implemented. (Win32)
1412 =item binmode FILEHANDLE
1414 Meaningless. (S<Mac OS>, S<RISC OS>)
1416 Reopens file and restores pointer; if function fails, underlying
1417 filehandle may be closed, or pointer may be in a different position.
1420 The value returned by C<tell> may be affected after the call, and
1421 the filehandle may be flushed. (Win32)
1425 Only limited meaning. Disabling/enabling write permission is mapped to
1426 locking/unlocking the file. (S<Mac OS>)
1428 Only good for changing "owner" read-write access, "group", and "other"
1429 bits are meaningless. (Win32)
1431 Only good for changing "owner" and "other" read-write access. (S<RISC OS>)
1433 Access permissions are mapped onto VOS access-control list changes. (VOS)
1437 Not implemented. (S<Mac OS>, Win32, Plan9, S<RISC OS>, VOS)
1439 Does nothing, but won't fail. (Win32)
1441 =item chroot FILENAME
1445 Not implemented. (S<Mac OS>, Win32, VMS, Plan9, S<RISC OS>, VOS, VM/ESA)
1447 =item crypt PLAINTEXT,SALT
1449 May not be available if library or source was not provided when building
1452 Not implemented. (VOS)
1456 Not implemented. (VMS, Plan9, VOS)
1458 =item dbmopen HASH,DBNAME,MODE
1460 Not implemented. (VMS, Plan9, VOS)
1464 Not useful. (S<Mac OS>, S<RISC OS>)
1466 Not implemented. (Win32)
1468 Invokes VMS debugger. (VMS)
1472 Not implemented. (S<Mac OS>)
1474 Implemented via Spawn. (VM/ESA)
1476 Does not automatically flush output handles on some platforms.
1477 (SunOS, Solaris, HP-UX)
1483 Emulates UNIX exit() (which considers C<exit 1> to indicate an error) by
1484 mapping the C<1> to SS$_ABORT (C<44>). This behavior may be overridden
1485 with the pragma C<use vmsish 'exit'>. As with the CRTL's exit()
1486 function, C<exit 0> is also mapped to an exit status of SS$_NORMAL
1487 (C<1>); this mapping cannot be overridden. Any other argument to exit()
1488 is used directly as Perl's exit status. (VMS)
1490 =item fcntl FILEHANDLE,FUNCTION,SCALAR
1492 Not implemented. (Win32, VMS)
1494 =item flock FILEHANDLE,OPERATION
1496 Not implemented (S<Mac OS>, VMS, S<RISC OS>, VOS).
1498 Available only on Windows NT (not on Windows 95). (Win32)
1502 Not implemented. (S<Mac OS>, AmigaOS, S<RISC OS>, VOS, VM/ESA)
1504 Emulated using multiple interpreters. See L<perlfork>. (Win32)
1506 Does not automatically flush output handles on some platforms.
1507 (SunOS, Solaris, HP-UX)
1511 Not implemented. (S<Mac OS>, S<RISC OS>)
1515 Not implemented. (S<Mac OS>, Win32, VMS, S<RISC OS>, VOS)
1519 Not implemented. (S<Mac OS>, Win32, VMS, S<RISC OS>)
1521 =item getpriority WHICH,WHO
1523 Not implemented. (S<Mac OS>, Win32, VMS, S<RISC OS>, VOS, VM/ESA)
1527 Not implemented. (S<Mac OS>, Win32)
1529 Not useful. (S<RISC OS>)
1533 Not implemented. (S<Mac OS>, Win32, VMS, S<RISC OS>)
1535 =item getnetbyname NAME
1537 Not implemented. (S<Mac OS>, Win32, Plan9)
1541 Not implemented. (S<Mac OS>, Win32)
1543 Not useful. (S<RISC OS>)
1547 Not implemented. (S<Mac OS>, Win32, VMS, S<RISC OS>)
1549 =item getnetbyaddr ADDR,ADDRTYPE
1551 Not implemented. (S<Mac OS>, Win32, Plan9)
1553 =item getprotobynumber NUMBER
1555 Not implemented. (S<Mac OS>)
1557 =item getservbyport PORT,PROTO
1559 Not implemented. (S<Mac OS>)
1563 Not implemented. (S<Mac OS>, Win32, VM/ESA)
1567 Not implemented. (S<Mac OS>, Win32, VMS, VM/ESA)
1571 Not implemented. (S<Mac OS>, Win32)
1575 Not implemented. (S<Mac OS>, Win32, Plan9)
1579 Not implemented. (S<Mac OS>, Win32, Plan9)
1583 Not implemented. (Win32, Plan9)
1587 Not implemented. (S<Mac OS>, Win32, S<RISC OS>)
1591 Not implemented. (S<Mac OS>, Win32, VMS, S<RISC OS>)
1593 =item sethostent STAYOPEN
1595 Not implemented. (S<Mac OS>, Win32, Plan9, S<RISC OS>)
1597 =item setnetent STAYOPEN
1599 Not implemented. (S<Mac OS>, Win32, Plan9, S<RISC OS>)
1601 =item setprotoent STAYOPEN
1603 Not implemented. (S<Mac OS>, Win32, Plan9, S<RISC OS>)
1605 =item setservent STAYOPEN
1607 Not implemented. (Plan9, Win32, S<RISC OS>)
1611 Not implemented. (S<Mac OS>, MPE/iX, VM/ESA, Win32)
1615 Not implemented. (S<Mac OS>, MPE/iX, S<RISC OS>, VM/ESA, VMS, Win32)
1619 Not implemented. (S<Mac OS>, Win32)
1623 Not implemented. (S<Mac OS>, Win32, Plan9)
1627 Not implemented. (S<Mac OS>, Win32, Plan9)
1631 Not implemented. (Plan9, Win32)
1633 =item getsockopt SOCKET,LEVEL,OPTNAME
1635 Not implemented. (Plan9)
1641 This operator is implemented via the File::Glob extension on most
1642 platforms. See L<File::Glob> for portability information.
1644 =item ioctl FILEHANDLE,FUNCTION,SCALAR
1646 Not implemented. (VMS)
1648 Available only for socket handles, and it does what the ioctlsocket() call
1649 in the Winsock API does. (Win32)
1651 Available only for socket handles. (S<RISC OS>)
1653 =item kill SIGNAL, LIST
1655 C<kill(0, LIST)> is implemented for the sake of taint checking;
1656 use with other signals is unimplemented. (S<Mac OS>)
1658 Not implemented, hence not useful for taint checking. (S<RISC OS>)
1660 C<kill()> doesn't have the semantics of C<raise()>, i.e. it doesn't send
1661 a signal to the identified process like it does on Unix platforms.
1662 Instead C<kill($sig, $pid)> terminates the process identified by $pid,
1663 and makes it exit immediately with exit status $sig. As in Unix, if
1664 $sig is 0 and the specified process exists, it returns true without
1665 actually terminating it. (Win32)
1667 =item link OLDFILE,NEWFILE
1669 Not implemented. (S<Mac OS>, MPE/iX, VMS, S<RISC OS>)
1671 Link count not updated because hard links are not quite that hard
1672 (They are sort of half-way between hard and soft links). (AmigaOS)
1674 Hard links are implemented on Win32 (Windows NT and Windows 2000)
1677 =item lstat FILEHANDLE
1683 Not implemented. (VMS, S<RISC OS>)
1685 Return values (especially for device and inode) may be bogus. (Win32)
1687 =item msgctl ID,CMD,ARG
1689 =item msgget KEY,FLAGS
1691 =item msgsnd ID,MSG,FLAGS
1693 =item msgrcv ID,VAR,SIZE,TYPE,FLAGS
1695 Not implemented. (S<Mac OS>, Win32, VMS, Plan9, S<RISC OS>, VOS)
1697 =item open FILEHANDLE,EXPR
1699 =item open FILEHANDLE
1701 The C<|> variants are supported only if ToolServer is installed.
1704 open to C<|-> and C<-|> are unsupported. (S<Mac OS>, Win32, S<RISC OS>)
1706 Opening a process does not automatically flush output handles on some
1707 platforms. (SunOS, Solaris, HP-UX)
1709 =item pipe READHANDLE,WRITEHANDLE
1711 Very limited functionality. (MiNT)
1717 Not implemented. (Win32, VMS, S<RISC OS>)
1719 =item select RBITS,WBITS,EBITS,TIMEOUT
1721 Only implemented on sockets. (Win32, VMS)
1723 Only reliable on sockets. (S<RISC OS>)
1725 Note that the C<select FILEHANDLE> form is generally portable.
1727 =item semctl ID,SEMNUM,CMD,ARG
1729 =item semget KEY,NSEMS,FLAGS
1731 =item semop KEY,OPSTRING
1733 Not implemented. (S<Mac OS>, Win32, VMS, S<RISC OS>, VOS)
1737 Not implemented. (MPE/iX, Win32)
1739 =item setpgrp PID,PGRP
1741 Not implemented. (S<Mac OS>, Win32, VMS, S<RISC OS>, VOS)
1743 =item setpriority WHICH,WHO,PRIORITY
1745 Not implemented. (S<Mac OS>, Win32, VMS, S<RISC OS>, VOS)
1749 Not implemented. (MPE/iX, Win32)
1751 =item setsockopt SOCKET,LEVEL,OPTNAME,OPTVAL
1753 Not implemented. (Plan9)
1755 =item shmctl ID,CMD,ARG
1757 =item shmget KEY,SIZE,FLAGS
1759 =item shmread ID,VAR,POS,SIZE
1761 =item shmwrite ID,STRING,POS,SIZE
1763 Not implemented. (S<Mac OS>, Win32, VMS, S<RISC OS>, VOS)
1765 =item sockatmark SOCKET
1767 A relatively recent addition to socket functions, may not
1768 be implemented even in UNIX platforms.
1770 =item socketpair SOCKET1,SOCKET2,DOMAIN,TYPE,PROTOCOL
1772 Not implemented. (Win32, VMS, S<RISC OS>, VOS, VM/ESA)
1774 =item stat FILEHANDLE
1780 Platforms that do not have rdev, blksize, or blocks will return these
1781 as '', so numeric comparison or manipulation of these fields may cause
1782 'not numeric' warnings.
1784 mtime and atime are the same thing, and ctime is creation time instead of
1785 inode change time. (S<Mac OS>)
1787 device and inode are not meaningful. (Win32)
1789 device and inode are not necessarily reliable. (VMS)
1791 mtime, atime and ctime all return the last modification time. Device and
1792 inode are not necessarily reliable. (S<RISC OS>)
1794 dev, rdev, blksize, and blocks are not available. inode is not
1795 meaningful and will differ between stat calls on the same file. (os2)
1797 some versions of cygwin when doing a stat("foo") and if not finding it
1798 may then attempt to stat("foo.exe") (Cygwin)
1800 =item symlink OLDFILE,NEWFILE
1802 Not implemented. (Win32, VMS, S<RISC OS>)
1806 Not implemented. (S<Mac OS>, Win32, VMS, S<RISC OS>, VOS, VM/ESA)
1808 =item sysopen FILEHANDLE,FILENAME,MODE,PERMS
1810 The traditional "0", "1", and "2" MODEs are implemented with different
1811 numeric values on some systems. The flags exported by C<Fcntl>
1812 (O_RDONLY, O_WRONLY, O_RDWR) should work everywhere though. (S<Mac
1813 OS>, OS/390, VM/ESA)
1817 In general, do not assume the UNIX/POSIX semantics that you can shift
1818 C<$?> right by eight to get the exit value, or that C<$? & 127>
1819 would give you the number of the signal that terminated the program,
1820 or that C<$? & 128> would test true if the program was terminated by a
1821 coredump. Instead, use the POSIX W*() interfaces: for example, use
1822 WIFEXITED($?) an WEXITVALUE($?) to test for a normal exit and the exit
1823 value, and WIFSIGNALED($?) and WTERMSIG($?) for a signal exit and the
1824 signal. Core dumping is not a portable concept, so there's no portable
1825 way to test for that.
1827 Only implemented if ToolServer is installed. (S<Mac OS>)
1829 As an optimization, may not call the command shell specified in
1830 C<$ENV{PERL5SHELL}>. C<system(1, @args)> spawns an external
1831 process and immediately returns its process designator, without
1832 waiting for it to terminate. Return value may be used subsequently
1833 in C<wait> or C<waitpid>. Failure to spawn() a subprocess is indicated
1834 by setting $? to "255 << 8". C<$?> is set in a way compatible with
1835 Unix (i.e. the exitstatus of the subprocess is obtained by "$? >> 8",
1836 as described in the documentation). (Win32)
1838 There is no shell to process metacharacters, and the native standard is
1839 to pass a command line terminated by "\n" "\r" or "\0" to the spawned
1840 program. Redirection such as C<< > foo >> is performed (if at all) by
1841 the run time library of the spawned program. C<system> I<list> will call
1842 the Unix emulation library's C<exec> emulation, which attempts to provide
1843 emulation of the stdin, stdout, stderr in force in the parent, providing
1844 the child program uses a compatible version of the emulation library.
1845 I<scalar> will call the native command line direct and no such emulation
1846 of a child Unix program will exists. Mileage B<will> vary. (S<RISC OS>)
1848 Far from being POSIX compliant. Because there may be no underlying
1849 /bin/sh tries to work around the problem by forking and execing the
1850 first token in its argument string. Handles basic redirection
1851 ("<" or ">") on its own behalf. (MiNT)
1853 Does not automatically flush output handles on some platforms.
1854 (SunOS, Solaris, HP-UX)
1856 The return value is POSIX-like (shifted up by 8 bits), which only allows
1857 room for a made-up value derived from the severity bits of the native
1858 32-bit condition code (unless overridden by C<use vmsish 'status'>).
1859 For more details see L<perlvms/$?>. (VMS)
1863 Only the first entry returned is nonzero. (S<Mac OS>)
1865 "cumulative" times will be bogus. On anything other than Windows NT
1866 or Windows 2000, "system" time will be bogus, and "user" time is
1867 actually the time returned by the clock() function in the C runtime
1870 Not useful. (S<RISC OS>)
1872 =item truncate FILEHANDLE,LENGTH
1874 =item truncate EXPR,LENGTH
1876 Not implemented. (Older versions of VMS)
1878 Truncation to zero-length only. (VOS)
1880 If a FILEHANDLE is supplied, it must be writable and opened in append
1881 mode (i.e., use C<<< open(FH, '>>filename') >>>
1882 or C<sysopen(FH,...,O_APPEND|O_RDWR)>. If a filename is supplied, it
1883 should not be held open elsewhere. (Win32)
1889 Returns undef where unavailable, as of version 5.005.
1891 C<umask> works but the correct permissions are set only when the file
1892 is finally closed. (AmigaOS)
1896 Only the modification time is updated. (S<BeOS>, S<Mac OS>, VMS, S<RISC OS>)
1898 May not behave as expected. Behavior depends on the C runtime
1899 library's implementation of utime(), and the filesystem being
1900 used. The FAT filesystem typically does not support an "access
1901 time" field, and it may limit timestamps to a granularity of
1902 two seconds. (Win32)
1906 =item waitpid PID,FLAGS
1908 Not implemented. (S<Mac OS>, VOS)
1910 Can only be applied to process handles returned for processes spawned
1911 using C<system(1, ...)> or pseudo processes created with C<fork()>. (Win32)
1913 Not useful. (S<RISC OS>)
1921 =item v1.48, 02 February 2001
1923 Various updates from perl5-porters over the past year, supported
1924 platforms update from Jarkko Hietaniemi.
1926 =item v1.47, 22 March 2000
1928 Various cleanups from Tom Christiansen, including migration of
1929 long platform listings from L<perl>.
1931 =item v1.46, 12 February 2000
1933 Updates for VOS and MPE/iX. (Peter Prymmer) Other small changes.
1935 =item v1.45, 20 December 1999
1937 Small changes from 5.005_63 distribution, more changes to EBCDIC info.
1939 =item v1.44, 19 July 1999
1941 A bunch of updates from Peter Prymmer for C<$^O> values,
1942 endianness, File::Spec, VMS, BS2000, OS/400.
1944 =item v1.43, 24 May 1999
1946 Added a lot of cleaning up from Tom Christiansen.
1948 =item v1.42, 22 May 1999
1950 Added notes about tests, sprintf/printf, and epoch offsets.
1952 =item v1.41, 19 May 1999
1954 Lots more little changes to formatting and content.
1956 Added a bunch of C<$^O> and related values
1957 for various platforms; fixed mail and web addresses, and added
1958 and changed miscellaneous notes. (Peter Prymmer)
1960 =item v1.40, 11 April 1999
1962 Miscellaneous changes.
1964 =item v1.39, 11 February 1999
1966 Changes from Jarkko and EMX URL fixes Michael Schwern. Additional
1967 note about newlines added.
1969 =item v1.38, 31 December 1998
1971 More changes from Jarkko.
1973 =item v1.37, 19 December 1998
1975 More minor changes. Merge two separate version 1.35 documents.
1977 =item v1.36, 9 September 1998
1979 Updated for Stratus VOS. Also known as version 1.35.
1981 =item v1.35, 13 August 1998
1983 Integrate more minor changes, plus addition of new sections under
1984 L<"ISSUES">: L<"Numbers endianness and Width">,
1985 L<"Character sets and character encoding">,
1986 L<"Internationalisation">.
1988 =item v1.33, 06 August 1998
1990 Integrate more minor changes.
1992 =item v1.32, 05 August 1998
1994 Integrate more minor changes.
1996 =item v1.30, 03 August 1998
1998 Major update for RISC OS, other minor changes.
2000 =item v1.23, 10 July 1998
2002 First public release with perl5.005.
2006 =head1 Supported Platforms
2008 As of early 2001 (the Perl releases 5.6.1 and 5.7.1), the following
2009 platforms are able to build Perl from the standard source code
2010 distribution available at http://www.cpan.org/src/index.html
2033 Tru64 UNIX (DEC OSF/1, Digital UNIX)
2039 1) in DOS mode either the DOS or OS/2 ports can be used
2040 2) Mac OS Classic (pre-X) is almost 5.6.1-ready; building from
2041 the source does work with 5.6.1, but additional MacOS specific
2042 source code is needed for a complete build. See the web
2043 site http://dev.macperl.org/ for more information.
2044 3) compilers: Borland, Cygwin, Mingw32 EGCS/GCC, VC++
2046 The following platforms worked for the previous releases (5.6.0 and 5.7.0),
2047 but we did not manage to test these in time for the 5.7.1 release.
2048 There is a very good chance that these will work fine with the 5.7.1.
2066 The following platform worked for the 5.005_03 major release but not
2067 for 5.6.0. Standardization on UTF-8 as the internal string
2068 representation in 5.6.0 and 5.6.1 introduced incompatibilities in this
2069 EBCDIC platform. While Perl 5.7.1 will build on this platform some
2070 regression tests may fail and the C<use utf8;> pragma typically
2071 introduces text handling errors.
2075 1) previously known as MVS, about to become z/OS.
2077 Strongly related to the OS/390 platform by also being EBCDIC-based
2078 mainframe platforms are the following platforms:
2083 These are also expected to work, albeit with no UTF-8 support, under 5.6.1
2084 for the same reasons as OS/390. Contact the mailing list perl-mvs@perl.org
2087 The following platforms have been known to build Perl from source in
2088 the past (5.005_03 and earlier), but we haven't been able to verify
2089 their status for the current release, either because the
2090 hardware/software platforms are rare or because we don't have an
2091 active champion on these platforms--or both. They used to work,
2092 though, so go ahead and try compiling them, and let perlbug@perl.org
2131 Support for the following platform is planned for a future Perl release:
2135 The following platforms have their own source code distributions and
2136 binaries available via http://www.cpan.org/ports/index.html:
2142 Tandem Guardian 5.004
2144 The following platforms have only binaries available via
2145 http://www.cpan.org/ports/index.html :
2149 Acorn RISCOS 5.005_02
2153 Although we do suggest that you always build your own Perl from
2154 the source code, both for maximal configurability and for security,
2155 in case you are in a hurry you can check
2156 http://www.cpan.org/ports/index.html for binary distributions.
2160 L<perlaix>, L<perlapollo>, L<perlamiga>, L<perlbeos>, L<perlbs200>,
2161 L<perlce>, L<perlcygwin>, L<perldgux>, L<perldos>, L<perlepoc>, L<perlebcdic>,
2162 L<perlhurd>, L<perlhpux>, L<perlmachten>, L<perlmacos>, L<perlmint>,
2163 L<perlmpeix>, L<perlnetware>, L<perlos2>, L<perlos390>, L<perlplan9>,
2164 L<perlqnx>, L<perlsolaris>, L<perltru64>, L<perlunicode>,
2165 L<perlvmesa>, L<perlvms>, L<perlvos>, L<perlwin32>, and L<Win32>.
2167 =head1 AUTHORS / CONTRIBUTORS
2169 Abigail <abigail@foad.org>,
2170 Charles Bailey <bailey@newman.upenn.edu>,
2171 Graham Barr <gbarr@pobox.com>,
2172 Tom Christiansen <tchrist@perl.com>,
2173 Nicholas Clark <nick@ccl4.org>,
2174 Thomas Dorner <Thomas.Dorner@start.de>,
2175 Andy Dougherty <doughera@lafayette.edu>,
2176 Dominic Dunlop <domo@computer.org>,
2177 Neale Ferguson <neale@vma.tabnsw.com.au>,
2178 David J. Fiander <davidf@mks.com>,
2179 Paul Green <Paul_Green@stratus.com>,
2180 M.J.T. Guy <mjtg@cam.ac.uk>,
2181 Jarkko Hietaniemi <jhi@iki.fi>,
2182 Luther Huffman <lutherh@stratcom.com>,
2183 Nick Ing-Simmons <nick@ing-simmons.net>,
2184 Andreas J. KE<ouml>nig <a.koenig@mind.de>,
2185 Markus Laker <mlaker@contax.co.uk>,
2186 Andrew M. Langmead <aml@world.std.com>,
2187 Larry Moore <ljmoore@freespace.net>,
2188 Paul Moore <Paul.Moore@uk.origin-it.com>,
2189 Chris Nandor <pudge@pobox.com>,
2190 Matthias Neeracher <neeri@iis.ee.ethz.ch>,
2191 Philip Newton <pne@cpan.org>,
2192 Gary Ng <71564.1743@CompuServe.COM>,
2193 Tom Phoenix <rootbeer@teleport.com>,
2194 AndrE<eacute> Pirard <A.Pirard@ulg.ac.be>,
2195 Peter Prymmer <pvhp@forte.com>,
2196 Hugo van der Sanden <hv@crypt0.demon.co.uk>,
2197 Gurusamy Sarathy <gsar@activestate.com>,
2198 Paul J. Schinder <schinder@pobox.com>,
2199 Michael G Schwern <schwern@pobox.com>,
2200 Dan Sugalski <dan@sidhe.org>,
2201 Nathan Torkington <gnat@frii.com>.
2205 Version 1.50, last modified 10 Jul 2001