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 (C<< <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 The "inode change timestamp" (the <-C> filetest) may really be the
266 "creation timestamp" (which it is not in UNIX).
268 VOS perl can emulate Unix filenames with C</> as path separator. The
269 native pathname characters greater-than, less-than, number-sign, and
270 percent-sign are always accepted.
272 S<RISC OS> perl can emulate Unix filenames with C</> as path
273 separator, or go native and use C<.> for path separator and C<:> to
274 signal filesystems and disk names.
276 Don't assume UNIX filesystem access semantics: that read, write,
277 and execute are all the permissions there are, and even if they exist,
278 that their semantics (for example what do r, w, and x mean on
279 a directory) are the UNIX ones. The various UNIX/POSIX compatibility
280 layers usually try to make interfaces like chmod() work, but sometimes
281 there simply is no good mapping.
283 If all this is intimidating, have no (well, maybe only a little)
284 fear. There are modules that can help. The File::Spec modules
285 provide methods to do the Right Thing on whatever platform happens
286 to be running the program.
288 use File::Spec::Functions;
289 chdir(updir()); # go up one directory
290 $file = catfile(curdir(), 'temp', 'file.txt');
291 # on Unix and Win32, './temp/file.txt'
292 # on Mac OS, ':temp:file.txt'
293 # on VMS, '[.temp]file.txt'
295 File::Spec is available in the standard distribution as of version
296 5.004_05. File::Spec::Functions is only in File::Spec 0.7 and later,
297 and some versions of perl come with version 0.6. If File::Spec
298 is not updated to 0.7 or later, you must use the object-oriented
299 interface from File::Spec (or upgrade File::Spec).
301 In general, production code should not have file paths hardcoded.
302 Making them user-supplied or read from a configuration file is
303 better, keeping in mind that file path syntax varies on different
306 This is especially noticeable in scripts like Makefiles and test suites,
307 which often assume C</> as a path separator for subdirectories.
309 Also of use is File::Basename from the standard distribution, which
310 splits a pathname into pieces (base filename, full path to directory,
313 Even when on a single platform (if you can call Unix a single platform),
314 remember not to count on the existence or the contents of particular
315 system-specific files or directories, like F</etc/passwd>,
316 F</etc/sendmail.conf>, F</etc/resolv.conf>, or even F</tmp/>. For
317 example, F</etc/passwd> may exist but not contain the encrypted
318 passwords, because the system is using some form of enhanced security.
319 Or it may not contain all the accounts, because the system is using NIS.
320 If code does need to rely on such a file, include a description of the
321 file and its format in the code's documentation, then make it easy for
322 the user to override the default location of the file.
324 Don't assume a text file will end with a newline. They should,
327 Do not have two files or directories of the same name with different
328 case, like F<test.pl> and F<Test.pl>, as many platforms have
329 case-insensitive (or at least case-forgiving) filenames. Also, try
330 not to have non-word characters (except for C<.>) in the names, and
331 keep them to the 8.3 convention, for maximum portability, onerous a
332 burden though this may appear.
334 Likewise, when using the AutoSplit module, try to keep your functions to
335 8.3 naming and case-insensitive conventions; or, at the least,
336 make it so the resulting files have a unique (case-insensitively)
339 Whitespace in filenames is tolerated on most systems, but not all,
340 and even on systems where it might be tolerated, some utilities
341 might become confused by such whitespace.
343 Many systems (DOS, VMS) cannot have more than one C<.> in their filenames.
345 Don't assume C<< > >> won't be the first character of a filename.
346 Always use C<< < >> explicitly to open a file for reading, or even
347 better, use the three-arg version of open, unless you want the user to
348 be able to specify a pipe open.
350 open(FILE, '<', $existing_file) or die $!;
352 If filenames might use strange characters, it is safest to open it
353 with C<sysopen> instead of C<open>. C<open> is magic and can
354 translate characters like C<< > >>, C<< < >>, and C<|>, which may
355 be the wrong thing to do. (Sometimes, though, it's the right thing.)
356 Three-arg open can also help protect against this translation in cases
357 where it is undesirable.
359 Don't use C<:> as a part of a filename since many systems use that for
360 their own semantics (MacOS Classic for separating pathname components,
361 many networking schemes and utilities for separating the nodename and
362 the pathname, and so on). For the same reasons, avoid C<@>, C<;> and
365 Don't assume that in pathnames you can collapse two leading slashes
366 C<//> into one: some networking and clustering filesystems have special
367 semantics for that. Let the operating system to sort it out.
369 The I<portable filename characters> as defined by ANSI C are
371 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
372 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
376 and the "-" shouldn't be the first character. If you want to be
377 hypercorrect, stay case-insensitive and within the 8.3 naming
378 convention (all the files and directories have to be unique within one
379 directory if their names are lowercased and truncated to eight
380 characters before the C<.>, if any, and to three characters after the
381 C<.>, if any). (And do not use C<.>s in directory names.)
383 =head2 System Interaction
385 Not all platforms provide a command line. These are usually platforms
386 that rely primarily on a Graphical User Interface (GUI) for user
387 interaction. A program requiring a command line interface might
388 not work everywhere. This is probably for the user of the program
389 to deal with, so don't stay up late worrying about it.
391 Some platforms can't delete or rename files held open by the system.
392 Remember to C<close> files when you are done with them. Don't
393 C<unlink> or C<rename> an open file. Don't C<tie> or C<open> a
394 file already tied or opened; C<untie> or C<close> it first.
396 Don't open the same file more than once at a time for writing, as some
397 operating systems put mandatory locks on such files.
399 Don't assume that write/modify permission on a directory gives the
400 right to add or delete files/directories in that directory. That is
401 filesystem specific: in some filesystems you need write/modify
402 permission also (or even just) in the file/directory itself. In some
403 filesystems (AFS, DFS) the permission to add/delete directory entries
404 is a completely separate permission.
406 Don't assume that a single C<unlink> completely gets rid of the file:
407 some filesystems (most notably the ones in VMS) have versioned
408 filesystems, and unlink() removes only the most recent one (it doesn't
409 remove all the versions because by default the native tools on those
410 platforms remove just the most recent version, too). The portable
411 idiom to remove all the versions of a file is
413 1 while unlink "file";
415 This will terminate if the file is undeleteable for some reason
416 (protected, not there, and so on).
418 Don't count on a specific environment variable existing in C<%ENV>.
419 Don't count on C<%ENV> entries being case-sensitive, or even
420 case-preserving. Don't try to clear %ENV by saying C<%ENV = ();>, or,
421 if you really have to, make it conditional on C<$^O ne 'VMS'> since in
422 VMS the C<%ENV> table is much more than a per-process key-value string
425 Don't count on signals or C<%SIG> for anything.
427 Don't count on filename globbing. Use C<opendir>, C<readdir>, and
430 Don't count on per-program environment variables, or per-program current
433 Don't count on specific values of C<$!>.
435 =head2 Interprocess Communication (IPC)
437 In general, don't directly access the system in code meant to be
438 portable. That means, no C<system>, C<exec>, C<fork>, C<pipe>,
439 C<``>, C<qx//>, C<open> with a C<|>, nor any of the other things
440 that makes being a perl hacker worth being.
442 Commands that launch external processes are generally supported on
443 most platforms (though many of them do not support any type of
444 forking). The problem with using them arises from what you invoke
445 them on. External tools are often named differently on different
446 platforms, may not be available in the same location, might accept
447 different arguments, can behave differently, and often present their
448 results in a platform-dependent way. Thus, you should seldom depend
449 on them to produce consistent results. (Then again, if you're calling
450 I<netstat -a>, you probably don't expect it to run on both Unix and CP/M.)
452 One especially common bit of Perl code is opening a pipe to B<sendmail>:
454 open(MAIL, '|/usr/lib/sendmail -t')
455 or die "cannot fork sendmail: $!";
457 This is fine for systems programming when sendmail is known to be
458 available. But it is not fine for many non-Unix systems, and even
459 some Unix systems that may not have sendmail installed. If a portable
460 solution is needed, see the various distributions on CPAN that deal
461 with it. Mail::Mailer and Mail::Send in the MailTools distribution are
462 commonly used, and provide several mailing methods, including mail,
463 sendmail, and direct SMTP (via Net::SMTP) if a mail transfer agent is
464 not available. Mail::Sendmail is a standalone module that provides
465 simple, platform-independent mailing.
467 The Unix System V IPC (C<msg*(), sem*(), shm*()>) is not available
468 even on all Unix platforms.
470 Do not use either the bare result of C<pack("N", 10, 20, 30, 40)> or
471 bare v-strings (such as C<v10.20.30.40>) to represent IPv4 addresses:
472 both forms just pack the four bytes into network order. That this
473 would be equal to the C language C<in_addr> struct (which is what the
474 socket code internally uses) is not guaranteed. To be portable use
475 the routines of the Socket extension, such as C<inet_aton()>,
476 C<inet_ntoa()>, and C<sockaddr_in()>.
478 The rule of thumb for portable code is: Do it all in portable Perl, or
479 use a module (that may internally implement it with platform-specific
480 code, but expose a common interface).
482 =head2 External Subroutines (XS)
484 XS code can usually be made to work with any platform, but dependent
485 libraries, header files, etc., might not be readily available or
486 portable, or the XS code itself might be platform-specific, just as Perl
487 code might be. If the libraries and headers are portable, then it is
488 normally reasonable to make sure the XS code is portable, too.
490 A different type of portability issue arises when writing XS code:
491 availability of a C compiler on the end-user's system. C brings
492 with it its own portability issues, and writing XS code will expose
493 you to some of those. Writing purely in Perl is an easier way to
496 =head2 Standard Modules
498 In general, the standard modules work across platforms. Notable
499 exceptions are the CPAN module (which currently makes connections to external
500 programs that may not be available), platform-specific modules (like
501 ExtUtils::MM_VMS), and DBM modules.
503 There is no one DBM module available on all platforms.
504 SDBM_File and the others are generally available on all Unix and DOSish
505 ports, but not in MacPerl, where only NBDM_File and DB_File are
508 The good news is that at least some DBM module should be available, and
509 AnyDBM_File will use whichever module it can find. Of course, then
510 the code needs to be fairly strict, dropping to the greatest common
511 factor (e.g., not exceeding 1K for each record), so that it will
512 work with any DBM module. See L<AnyDBM_File> for more details.
516 The system's notion of time of day and calendar date is controlled in
517 widely different ways. Don't assume the timezone is stored in C<$ENV{TZ}>,
518 and even if it is, don't assume that you can control the timezone through
521 Don't assume that the epoch starts at 00:00:00, January 1, 1970,
522 because that is OS- and implementation-specific. It is better to store a date
523 in an unambiguous representation. The ISO-8601 standard defines
524 "YYYY-MM-DD" as the date format. A text representation (like "1987-12-18")
525 can be easily converted into an OS-specific value using a module like
526 Date::Parse. An array of values, such as those returned by
527 C<localtime>, can be converted to an OS-specific representation using
530 When calculating specific times, such as for tests in time or date modules,
531 it may be appropriate to calculate an offset for the epoch.
534 $offset = Time::Local::timegm(0, 0, 0, 1, 0, 70);
536 The value for C<$offset> in Unix will be C<0>, but in Mac OS will be
537 some large number. C<$offset> can then be added to a Unix time value
538 to get what should be the proper value on any system.
540 =head2 Character sets and character encoding
542 Assume very little about character sets.
544 Assume nothing about numerical values (C<ord>, C<chr>) of characters.
545 Do not use explicit code point ranges (like \xHH-\xHH); use for
546 example symbolic character classes like C<[:print:]>.
548 Do not assume that the alphabetic characters are encoded contiguously
549 (in the numeric sense). There may be gaps.
551 Do not assume anything about the ordering of the characters.
552 The lowercase letters may come before or after the uppercase letters;
553 the lowercase and uppercase may be interlaced so that both `a' and `A'
554 come before `b'; the accented and other international characters may
555 be interlaced so that E<auml> comes before `b'.
557 =head2 Internationalisation
559 If you may assume POSIX (a rather large assumption), you may read
560 more about the POSIX locale system from L<perllocale>. The locale
561 system at least attempts to make things a little bit more portable,
562 or at least more convenient and native-friendly for non-English
563 users. The system affects character sets and encoding, and date
564 and time formatting--amongst other things.
566 =head2 System Resources
568 If your code is destined for systems with severely constrained (or
569 missing!) virtual memory systems then you want to be I<especially> mindful
570 of avoiding wasteful constructs such as:
572 # NOTE: this is no longer "bad" in perl5.005
573 for (0..10000000) {} # bad
574 for (my $x = 0; $x <= 10000000; ++$x) {} # good
576 @lines = <VERY_LARGE_FILE>; # bad
578 while (<FILE>) {$file .= $_} # sometimes bad
579 $file = join('', <FILE>); # better
581 The last two constructs may appear unintuitive to most people. The
582 first repeatedly grows a string, whereas the second allocates a
583 large chunk of memory in one go. On some systems, the second is
584 more efficient that the first.
588 Most multi-user platforms provide basic levels of security, usually
589 implemented at the filesystem level. Some, however, do
590 not-- unfortunately. Thus the notion of user id, or "home" directory,
591 or even the state of being logged-in, may be unrecognizable on many
592 platforms. If you write programs that are security-conscious, it
593 is usually best to know what type of system you will be running
594 under so that you can write code explicitly for that platform (or
597 Don't assume the UNIX filesystem access semantics: the operating
598 system or the filesystem may be using some ACL systems, which are
599 richer languages than the usual rwx. Even if the rwx exist,
600 their semantics might be different.
602 (From security viewpoint testing for permissions before attempting to
603 do something is silly anyway: if one tries this, there is potential
604 for race conditions-- someone or something might change the
605 permissions between the permissions check and the actual operation.
606 Just try the operation.)
608 Don't assume the UNIX user and group semantics: especially, don't
609 expect the C<< $< >> and C<< $> >> (or the C<$(> and C<$)>) to work
610 for switching identities (or memberships).
612 Don't assume set-uid and set-gid semantics. (And even if you do,
613 think twice: set-uid and set-gid are a known can of security worms.)
617 For those times when it is necessary to have platform-specific code,
618 consider keeping the platform-specific code in one place, making porting
619 to other platforms easier. Use the Config module and the special
620 variable C<$^O> to differentiate platforms, as described in
623 Be careful in the tests you supply with your module or programs.
624 Module code may be fully portable, but its tests might not be. This
625 often happens when tests spawn off other processes or call external
626 programs to aid in the testing, or when (as noted above) the tests
627 assume certain things about the filesystem and paths. Be careful
628 not to depend on a specific output style for errors, such as when
629 checking C<$!> after a system call. Some platforms expect a certain
630 output format, and perl on those platforms may have been adjusted
631 accordingly. Most specifically, don't anchor a regex when testing
636 Modules uploaded to CPAN are tested by a variety of volunteers on
637 different platforms. These CPAN testers are notified by mail of each
638 new upload, and reply to the list with PASS, FAIL, NA (not applicable to
639 this platform), or UNKNOWN (unknown), along with any relevant notations.
641 The purpose of the testing is twofold: one, to help developers fix any
642 problems in their code that crop up because of lack of testing on other
643 platforms; two, to provide users with information about whether
644 a given module works on a given platform.
648 =item Mailing list: cpan-testers@perl.org
650 =item Testing results: http://testers.cpan.org/
656 As of version 5.002, Perl is built with a C<$^O> variable that
657 indicates the operating system it was built on. This was implemented
658 to help speed up code that would otherwise have to C<use Config>
659 and use the value of C<$Config{osname}>. Of course, to get more
660 detailed information about the system, looking into C<%Config> is
661 certainly recommended.
663 C<%Config> cannot always be trusted, however, because it was built
664 at compile time. If perl was built in one place, then transferred
665 elsewhere, some values may be wrong. The values may even have been
666 edited after the fact.
670 Perl works on a bewildering variety of Unix and Unix-like platforms (see
671 e.g. most of the files in the F<hints/> directory in the source code kit).
672 On most of these systems, the value of C<$^O> (hence C<$Config{'osname'}>,
673 too) is determined either by lowercasing and stripping punctuation from the
674 first field of the string returned by typing C<uname -a> (or a similar command)
675 at the shell prompt or by testing the file system for the presence of
676 uniquely named files such as a kernel or header file. Here, for example,
677 are a few of the more popular Unix flavors:
679 uname $^O $Config{'archname'}
680 --------------------------------------------
682 BSD/OS bsdos i386-bsdos
684 dgux dgux AViiON-dgux
685 DYNIX/ptx dynixptx i386-dynixptx
686 FreeBSD freebsd freebsd-i386
687 Linux linux arm-linux
688 Linux linux i386-linux
689 Linux linux i586-linux
690 Linux linux ppc-linux
691 HP-UX hpux PA-RISC1.1
693 Mac OS X darwin darwin
694 MachTen PPC machten powerpc-machten
696 NeXT 4 next OPENSTEP-Mach
697 openbsd openbsd i386-openbsd
698 OSF1 dec_osf alpha-dec_osf
699 reliantunix-n svr4 RM400-svr4
700 SCO_SV sco_sv i386-sco_sv
701 SINIX-N svr4 RM400-svr4
702 sn4609 unicos CRAY_C90-unicos
703 sn6521 unicosmk t3e-unicosmk
704 sn9617 unicos CRAY_J90-unicos
705 SunOS solaris sun4-solaris
706 SunOS solaris i86pc-solaris
707 SunOS4 sunos sun4-sunos
709 Because the value of C<$Config{archname}> may depend on the
710 hardware architecture, it can vary more than the value of C<$^O>.
712 =head2 DOS and Derivatives
714 Perl has long been ported to Intel-style microcomputers running under
715 systems like PC-DOS, MS-DOS, OS/2, and most Windows platforms you can
716 bring yourself to mention (except for Windows CE, if you count that).
717 Users familiar with I<COMMAND.COM> or I<CMD.EXE> style shells should
718 be aware that each of these file specifications may have subtle
721 $filespec0 = "c:/foo/bar/file.txt";
722 $filespec1 = "c:\\foo\\bar\\file.txt";
723 $filespec2 = 'c:\foo\bar\file.txt';
724 $filespec3 = 'c:\\foo\\bar\\file.txt';
726 System calls accept either C</> or C<\> as the path separator.
727 However, many command-line utilities of DOS vintage treat C</> as
728 the option prefix, so may get confused by filenames containing C</>.
729 Aside from calling any external programs, C</> will work just fine,
730 and probably better, as it is more consistent with popular usage,
731 and avoids the problem of remembering what to backwhack and what
734 The DOS FAT filesystem can accommodate only "8.3" style filenames. Under
735 the "case-insensitive, but case-preserving" HPFS (OS/2) and NTFS (NT)
736 filesystems you may have to be careful about case returned with functions
737 like C<readdir> or used with functions like C<open> or C<opendir>.
739 DOS also treats several filenames as special, such as AUX, PRN,
740 NUL, CON, COM1, LPT1, LPT2, etc. Unfortunately, sometimes these
741 filenames won't even work if you include an explicit directory
742 prefix. It is best to avoid such filenames, if you want your code
743 to be portable to DOS and its derivatives. It's hard to know what
744 these all are, unfortunately.
746 Users of these operating systems may also wish to make use of
747 scripts such as I<pl2bat.bat> or I<pl2cmd> to
748 put wrappers around your scripts.
750 Newline (C<\n>) is translated as C<\015\012> by STDIO when reading from
751 and writing to files (see L<"Newlines">). C<binmode(FILEHANDLE)>
752 will keep C<\n> translated as C<\012> for that filehandle. Since it is a
753 no-op on other systems, C<binmode> should be used for cross-platform code
754 that deals with binary data. That's assuming you realize in advance
755 that your data is in binary. General-purpose programs should
756 often assume nothing about their data.
758 The C<$^O> variable and the C<$Config{archname}> values for various
759 DOSish perls are as follows:
761 OS $^O $Config{archname} ID Version
762 --------------------------------------------------------
766 Windows 3.1 ? ? 0 3 01
767 Windows 95 MSWin32 MSWin32-x86 1 4 00
768 Windows 98 MSWin32 MSWin32-x86 1 4 10
769 Windows ME MSWin32 MSWin32-x86 1 ?
770 Windows NT MSWin32 MSWin32-x86 2 4 xx
771 Windows NT MSWin32 MSWin32-ALPHA 2 4 xx
772 Windows NT MSWin32 MSWin32-ppc 2 4 xx
773 Windows 2000 MSWin32 MSWin32-x86 2 5 xx
774 Windows XP MSWin32 MSWin32-x86 2 ?
775 Windows CE MSWin32 ? 3
778 The various MSWin32 Perl's can distinguish the OS they are running on
779 via the value of the fifth element of the list returned from
780 Win32::GetOSVersion(). For example:
782 if ($^O eq 'MSWin32') {
783 my @os_version_info = Win32::GetOSVersion();
784 print +('3.1','95','NT')[$os_version_info[4]],"\n";
793 The djgpp environment for DOS, http://www.delorie.com/djgpp/
798 The EMX environment for DOS, OS/2, etc. emx@iaehv.nl,
799 http://www.leo.org/pub/comp/os/os2/leo/gnu/emx+gcc/index.html or
800 ftp://hobbes.nmsu.edu/pub/os2/dev/emx/ Also L<perlos2>.
804 Build instructions for Win32 in L<perlwin32>, or under the Cygnus environment
809 The C<Win32::*> modules in L<Win32>.
813 The ActiveState Pages, http://www.activestate.com/
817 The Cygwin environment for Win32; F<README.cygwin> (installed
818 as L<perlcygwin>), http://www.cygwin.com/
822 The U/WIN environment for Win32,
823 http://www.research.att.com/sw/tools/uwin/
827 Build instructions for OS/2, L<perlos2>
833 Any module requiring XS compilation is right out for most people, because
834 MacPerl is built using non-free (and non-cheap!) compilers. Some XS
835 modules that can work with MacPerl are built and distributed in binary
838 Directories are specified as:
840 volume:folder:file for absolute pathnames
841 volume:folder: for absolute pathnames
842 :folder:file for relative pathnames
843 :folder: for relative pathnames
844 :file for relative pathnames
845 file for relative pathnames
847 Files are stored in the directory in alphabetical order. Filenames are
848 limited to 31 characters, and may include any character except for
849 null and C<:>, which is reserved as the path separator.
851 Instead of C<flock>, see C<FSpSetFLock> and C<FSpRstFLock> in the
852 Mac::Files module, or C<chmod(0444, ...)> and C<chmod(0666, ...)>.
854 In the MacPerl application, you can't run a program from the command line;
855 programs that expect C<@ARGV> to be populated can be edited with something
856 like the following, which brings up a dialog box asking for the command
860 @ARGV = split /\s+/, MacPerl::Ask('Arguments?');
863 A MacPerl script saved as a "droplet" will populate C<@ARGV> with the full
864 pathnames of the files dropped onto the script.
866 Mac users can run programs under a type of command line interface
867 under MPW (Macintosh Programmer's Workshop, a free development
868 environment from Apple). MacPerl was first introduced as an MPW
869 tool, and MPW can be used like a shell:
871 perl myscript.plx some arguments
873 ToolServer is another app from Apple that provides access to MPW tools
874 from MPW and the MacPerl app, which allows MacPerl programs to use
875 C<system>, backticks, and piped C<open>.
877 "S<Mac OS>" is the proper name for the operating system, but the value
878 in C<$^O> is "MacOS". To determine architecture, version, or whether
879 the application or MPW tool version is running, check:
881 $is_app = $MacPerl::Version =~ /App/;
882 $is_tool = $MacPerl::Version =~ /MPW/;
883 ($version) = $MacPerl::Version =~ /^(\S+)/;
884 $is_ppc = $MacPerl::Architecture eq 'MacPPC';
885 $is_68k = $MacPerl::Architecture eq 'Mac68K';
887 S<Mac OS X>, based on NeXT's OpenStep OS, runs MacPerl natively, under the
888 "Classic" environment. There is no "Carbon" version of MacPerl to run
889 under the primary Mac OS X environment. S<Mac OS X> and its Open Source
890 version, Darwin, both run Unix perl natively.
898 MacPerl Development, http://dev.macperl.org/ .
902 The MacPerl Pages, http://www.macperl.com/ .
906 The MacPerl mailing lists, http://lists.perl.org/ .
912 Perl on VMS is discussed in L<perlvms> in the perl distribution.
913 Perl on VMS can accept either VMS- or Unix-style file
914 specifications as in either of the following:
916 $ perl -ne "print if /perl_setup/i" SYS$LOGIN:LOGIN.COM
917 $ perl -ne "print if /perl_setup/i" /sys$login/login.com
919 but not a mixture of both as in:
921 $ perl -ne "print if /perl_setup/i" sys$login:/login.com
922 Can't open sys$login:/login.com: file specification syntax error
924 Interacting with Perl from the Digital Command Language (DCL) shell
925 often requires a different set of quotation marks than Unix shells do.
928 $ perl -e "print ""Hello, world.\n"""
931 There are several ways to wrap your perl scripts in DCL F<.COM> files, if
932 you are so inclined. For example:
934 $ write sys$output "Hello from DCL!"
936 $ then perl -x 'f$environment("PROCEDURE")
937 $ else perl -x - 'p1 'p2 'p3 'p4 'p5 'p6 'p7 'p8
938 $ deck/dollars="__END__"
941 print "Hello from Perl!\n";
946 Do take care with C<$ ASSIGN/nolog/user SYS$COMMAND: SYS$INPUT> if your
947 perl-in-DCL script expects to do things like C<< $read = <STDIN>; >>.
949 Filenames are in the format "name.extension;version". The maximum
950 length for filenames is 39 characters, and the maximum length for
951 extensions is also 39 characters. Version is a number from 1 to
952 32767. Valid characters are C</[A-Z0-9$_-]/>.
954 VMS's RMS filesystem is case-insensitive and does not preserve case.
955 C<readdir> returns lowercased filenames, but specifying a file for
956 opening remains case-insensitive. Files without extensions have a
957 trailing period on them, so doing a C<readdir> with a file named F<A.;5>
958 will return F<a.> (though that file could be opened with
961 RMS had an eight level limit on directory depths from any rooted logical
962 (allowing 16 levels overall) prior to VMS 7.2. Hence
963 C<PERL_ROOT:[LIB.2.3.4.5.6.7.8]> is a valid directory specification but
964 C<PERL_ROOT:[LIB.2.3.4.5.6.7.8.9]> is not. F<Makefile.PL> authors might
965 have to take this into account, but at least they can refer to the former
966 as C</PERL_ROOT/lib/2/3/4/5/6/7/8/>.
968 The VMS::Filespec module, which gets installed as part of the build
969 process on VMS, is a pure Perl module that can easily be installed on
970 non-VMS platforms and can be helpful for conversions to and from RMS
973 What C<\n> represents depends on the type of file opened. It usually
974 represents C<\012> but it could also be C<\015>, C<\012>, C<\015\012>,
975 C<\000>, C<\040>, or nothing depending on the file organiztion and
976 record format. The VMS::Stdio module provides access to the
977 special fopen() requirements of files with unusual attributes on VMS.
979 TCP/IP stacks are optional on VMS, so socket routines might not be
980 implemented. UDP sockets may not be supported.
982 The value of C<$^O> on OpenVMS is "VMS". To determine the architecture
983 that you are running on without resorting to loading all of C<%Config>
984 you can examine the content of the C<@INC> array like so:
986 if (grep(/VMS_AXP/, @INC)) {
987 print "I'm on Alpha!\n";
989 } elsif (grep(/VMS_VAX/, @INC)) {
990 print "I'm on VAX!\n";
993 print "I'm not so sure about where $^O is...\n";
996 On VMS, perl determines the UTC offset from the C<SYS$TIMEZONE_DIFFERENTIAL>
997 logical name. Although the VMS epoch began at 17-NOV-1858 00:00:00.00,
998 calls to C<localtime> are adjusted to count offsets from
999 01-JAN-1970 00:00:00.00, just like Unix.
1007 F<README.vms> (installed as L<README_vms>), L<perlvms>
1011 vmsperl list, majordomo@perl.org
1013 (Put the words C<subscribe vmsperl> in message body.)
1017 vmsperl on the web, http://www.sidhe.org/vmsperl/index.html
1023 Perl on VOS is discussed in F<README.vos> in the perl distribution
1024 (installed as L<perlvos>). Perl on VOS can accept either VOS- or
1025 Unix-style file specifications as in either of the following:
1027 C<< $ perl -ne "print if /perl_setup/i" >system>notices >>
1028 C<< $ perl -ne "print if /perl_setup/i" /system/notices >>
1030 or even a mixture of both as in:
1032 C<< $ perl -ne "print if /perl_setup/i" >system/notices >>
1034 Even though VOS allows the slash character to appear in object
1035 names, because the VOS port of Perl interprets it as a pathname
1036 delimiting character, VOS files, directories, or links whose names
1037 contain a slash character cannot be processed. Such files must be
1038 renamed before they can be processed by Perl. Note that VOS limits
1039 file names to 32 or fewer characters.
1041 Perl on VOS can be built using two different compilers and two different
1042 versions of the POSIX runtime. The recommended method for building full
1043 Perl is with the GNU C compiler and the generally-available version of
1044 VOS POSIX support. See F<README.vos> (installed as L<perlvos>) for
1045 restrictions that apply when Perl is built using the VOS Standard C
1046 compiler or the alpha version of VOS POSIX support.
1048 The value of C<$^O> on VOS is "VOS". To determine the architecture that
1049 you are running on without resorting to loading all of C<%Config> you
1050 can examine the content of the @INC array like so:
1053 print "I'm on a Stratus box!\n";
1055 print "I'm not on a Stratus box!\n";
1059 if (grep(/860/, @INC)) {
1060 print "This box is a Stratus XA/R!\n";
1062 } elsif (grep(/7100/, @INC)) {
1063 print "This box is a Stratus HP 7100 or 8xxx!\n";
1065 } elsif (grep(/8000/, @INC)) {
1066 print "This box is a Stratus HP 8xxx!\n";
1069 print "This box is a Stratus 68K!\n";
1078 F<README.vos> (installed as L<perlvos>)
1082 The VOS mailing list.
1084 There is no specific mailing list for Perl on VOS. You can post
1085 comments to the comp.sys.stratus newsgroup, or subscribe to the general
1086 Stratus mailing list. Send a letter with "subscribe Info-Stratus" in
1087 the message body to majordomo@list.stratagy.com.
1091 VOS Perl on the web at http://ftp.stratus.com/pub/vos/posix/posix.html
1095 =head2 EBCDIC Platforms
1097 Recent versions of Perl have been ported to platforms such as OS/400 on
1098 AS/400 minicomputers as well as OS/390, VM/ESA, and BS2000 for S/390
1099 Mainframes. Such computers use EBCDIC character sets internally (usually
1100 Character Code Set ID 0037 for OS/400 and either 1047 or POSIX-BC for S/390
1101 systems). On the mainframe perl currently works under the "Unix system
1102 services for OS/390" (formerly known as OpenEdition), VM/ESA OpenEdition, or
1103 the BS200 POSIX-BC system (BS2000 is supported in perl 5.6 and greater).
1104 See L<perlos390> for details.
1106 As of R2.5 of USS for OS/390 and Version 2.3 of VM/ESA these Unix
1107 sub-systems do not support the C<#!> shebang trick for script invocation.
1108 Hence, on OS/390 and VM/ESA perl scripts can be executed with a header
1109 similar to the following simple script:
1112 eval 'exec /usr/local/bin/perl -S $0 ${1+"$@"}'
1114 #!/usr/local/bin/perl # just a comment really
1116 print "Hello from perl!\n";
1118 OS/390 will support the C<#!> shebang trick in release 2.8 and beyond.
1119 Calls to C<system> and backticks can use POSIX shell syntax on all
1122 On the AS/400, if PERL5 is in your library list, you may need
1123 to wrap your perl scripts in a CL procedure to invoke them like so:
1126 CALL PGM(PERL5/PERL) PARM('/QOpenSys/hello.pl')
1129 This will invoke the perl script F<hello.pl> in the root of the
1130 QOpenSys file system. On the AS/400 calls to C<system> or backticks
1133 On these platforms, bear in mind that the EBCDIC character set may have
1134 an effect on what happens with some perl functions (such as C<chr>,
1135 C<pack>, C<print>, C<printf>, C<ord>, C<sort>, C<sprintf>, C<unpack>), as
1136 well as bit-fiddling with ASCII constants using operators like C<^>, C<&>
1137 and C<|>, not to mention dealing with socket interfaces to ASCII computers
1138 (see L<"Newlines">).
1140 Fortunately, most web servers for the mainframe will correctly
1141 translate the C<\n> in the following statement to its ASCII equivalent
1142 (C<\r> is the same under both Unix and OS/390 & VM/ESA):
1144 print "Content-type: text/html\r\n\r\n";
1146 The values of C<$^O> on some of these platforms includes:
1148 uname $^O $Config{'archname'}
1149 --------------------------------------------
1152 POSIX-BC posix-bc BS2000-posix-bc
1155 Some simple tricks for determining if you are running on an EBCDIC
1156 platform could include any of the following (perhaps all):
1158 if ("\t" eq "\05") { print "EBCDIC may be spoken here!\n"; }
1160 if (ord('A') == 193) { print "EBCDIC may be spoken here!\n"; }
1162 if (chr(169) eq 'z') { print "EBCDIC may be spoken here!\n"; }
1164 One thing you may not want to rely on is the EBCDIC encoding
1165 of punctuation characters since these may differ from code page to code
1166 page (and once your module or script is rumoured to work with EBCDIC,
1167 folks will want it to work with all EBCDIC character sets).
1177 L<perlos390>, F<README.os390>, F<perlbs2000>, F<README.vmesa>,
1182 The perl-mvs@perl.org list is for discussion of porting issues as well as
1183 general usage issues for all EBCDIC Perls. Send a message body of
1184 "subscribe perl-mvs" to majordomo@perl.org.
1188 AS/400 Perl information at
1189 http://as400.rochester.ibm.com/
1190 as well as on CPAN in the F<ports/> directory.
1194 =head2 Acorn RISC OS
1196 Because Acorns use ASCII with newlines (C<\n>) in text files as C<\012> like
1197 Unix, and because Unix filename emulation is turned on by default,
1198 most simple scripts will probably work "out of the box". The native
1199 filesystem is modular, and individual filesystems are free to be
1200 case-sensitive or insensitive, and are usually case-preserving. Some
1201 native filesystems have name length limits, which file and directory
1202 names are silently truncated to fit. Scripts should be aware that the
1203 standard filesystem currently has a name length limit of B<10>
1204 characters, with up to 77 items in a directory, but other filesystems
1205 may not impose such limitations.
1207 Native filenames are of the form
1209 Filesystem#Special_Field::DiskName.$.Directory.Directory.File
1213 Special_Field is not usually present, but may contain . and $ .
1214 Filesystem =~ m|[A-Za-z0-9_]|
1215 DsicName =~ m|[A-Za-z0-9_/]|
1216 $ represents the root directory
1217 . is the path separator
1218 @ is the current directory (per filesystem but machine global)
1219 ^ is the parent directory
1220 Directory and File =~ m|[^\0- "\.\$\%\&:\@\\^\|\177]+|
1222 The default filename translation is roughly C<tr|/.|./|;>
1224 Note that C<"ADFS::HardDisk.$.File" ne 'ADFS::HardDisk.$.File'> and that
1225 the second stage of C<$> interpolation in regular expressions will fall
1226 foul of the C<$.> if scripts are not careful.
1228 Logical paths specified by system variables containing comma-separated
1229 search lists are also allowed; hence C<System:Modules> is a valid
1230 filename, and the filesystem will prefix C<Modules> with each section of
1231 C<System$Path> until a name is made that points to an object on disk.
1232 Writing to a new file C<System:Modules> would be allowed only if
1233 C<System$Path> contains a single item list. The filesystem will also
1234 expand system variables in filenames if enclosed in angle brackets, so
1235 C<< <System$Dir>.Modules >> would look for the file
1236 S<C<$ENV{'System$Dir'} . 'Modules'>>. The obvious implication of this is
1237 that B<fully qualified filenames can start with C<< <> >>> and should
1238 be protected when C<open> is used for input.
1240 Because C<.> was in use as a directory separator and filenames could not
1241 be assumed to be unique after 10 characters, Acorn implemented the C
1242 compiler to strip the trailing C<.c> C<.h> C<.s> and C<.o> suffix from
1243 filenames specified in source code and store the respective files in
1244 subdirectories named after the suffix. Hence files are translated:
1247 C:foo.h C:h.foo (logical path variable)
1248 sys/os.h sys.h.os (C compiler groks Unix-speak)
1249 10charname.c c.10charname
1250 10charname.o o.10charname
1251 11charname_.c c.11charname (assuming filesystem truncates at 10)
1253 The Unix emulation library's translation of filenames to native assumes
1254 that this sort of translation is required, and it allows a user-defined list
1255 of known suffixes that it will transpose in this fashion. This may
1256 seem transparent, but consider that with these rules C<foo/bar/baz.h>
1257 and C<foo/bar/h/baz> both map to C<foo.bar.h.baz>, and that C<readdir> and
1258 C<glob> cannot and do not attempt to emulate the reverse mapping. Other
1259 C<.>'s in filenames are translated to C</>.
1261 As implied above, the environment accessed through C<%ENV> is global, and
1262 the convention is that program specific environment variables are of the
1263 form C<Program$Name>. Each filesystem maintains a current directory,
1264 and the current filesystem's current directory is the B<global> current
1265 directory. Consequently, sociable programs don't change the current
1266 directory but rely on full pathnames, and programs (and Makefiles) cannot
1267 assume that they can spawn a child process which can change the current
1268 directory without affecting its parent (and everyone else for that
1271 Because native operating system filehandles are global and are currently
1272 allocated down from 255, with 0 being a reserved value, the Unix emulation
1273 library emulates Unix filehandles. Consequently, you can't rely on
1274 passing C<STDIN>, C<STDOUT>, or C<STDERR> to your children.
1276 The desire of users to express filenames of the form
1277 C<< <Foo$Dir>.Bar >> on the command line unquoted causes problems,
1278 too: C<``> command output capture has to perform a guessing game. It
1279 assumes that a string C<< <[^<>]+\$[^<>]> >> is a
1280 reference to an environment variable, whereas anything else involving
1281 C<< < >> or C<< > >> is redirection, and generally manages to be 99%
1282 right. Of course, the problem remains that scripts cannot rely on any
1283 Unix tools being available, or that any tools found have Unix-like command
1286 Extensions and XS are, in theory, buildable by anyone using free
1287 tools. In practice, many don't, as users of the Acorn platform are
1288 used to binary distributions. MakeMaker does run, but no available
1289 make currently copes with MakeMaker's makefiles; even if and when
1290 this should be fixed, the lack of a Unix-like shell will cause
1291 problems with makefile rules, especially lines of the form C<cd
1292 sdbm && make all>, and anything using quoting.
1294 "S<RISC OS>" is the proper name for the operating system, but the value
1295 in C<$^O> is "riscos" (because we don't like shouting).
1299 Perl has been ported to many platforms that do not fit into any of
1300 the categories listed above. Some, such as AmigaOS, Atari MiNT,
1301 BeOS, HP MPE/iX, QNX, Plan 9, and VOS, have been well-integrated
1302 into the standard Perl source code kit. You may need to see the
1303 F<ports/> directory on CPAN for information, and possibly binaries,
1304 for the likes of: aos, Atari ST, lynxos, riscos, Novell Netware,
1305 Tandem Guardian, I<etc.> (Yes, we know that some of these OSes may
1306 fall under the Unix category, but we are not a standards body.)
1308 Some approximate operating system names and their C<$^O> values
1309 in the "OTHER" category include:
1311 OS $^O $Config{'archname'}
1312 ------------------------------------------
1313 Amiga DOS amigaos m68k-amigos
1314 MPE/iX mpeix PA-RISC1.1
1322 Amiga, F<README.amiga> (installed as L<perlamiga>).
1326 Atari, F<README.mint> and Guido Flohr's web page
1327 http://stud.uni-sb.de/~gufl0000/
1331 Be OS, F<README.beos>
1335 HP 300 MPE/iX, F<README.mpeix> and Mark Bixby's web page
1336 http://www.bixby.org/mark/perlix.html
1340 A free perl5-based PERL.NLM for Novell Netware is available in
1341 precompiled binary and source code form from http://www.novell.com/
1342 as well as from CPAN.
1346 Plan 9, F<README.plan9>
1350 =head1 FUNCTION IMPLEMENTATIONS
1352 Listed below are functions that are either completely unimplemented
1353 or else have been implemented differently on various platforms.
1354 Following each description will be, in parentheses, a list of
1355 platforms that the description applies to.
1357 The list may well be incomplete, or even wrong in some places. When
1358 in doubt, consult the platform-specific README files in the Perl
1359 source distribution, and any other documentation resources accompanying
1362 Be aware, moreover, that even among Unix-ish systems there are variations.
1364 For many functions, you can also query C<%Config>, exported by
1365 default from the Config module. For example, to check whether the
1366 platform has the C<lstat> call, check C<$Config{d_lstat}>. See
1367 L<Config> for a full description of available variables.
1369 =head2 Alphabetical Listing of Perl Functions
1379 C<-r>, C<-w>, and C<-x> have a limited meaning only; directories
1380 and applications are executable, and there are no uid/gid
1381 considerations. C<-o> is not supported. (S<Mac OS>)
1383 C<-r>, C<-w>, C<-x>, and C<-o> tell whether the file is accessible,
1384 which may not reflect UIC-based file protections. (VMS)
1386 C<-s> returns the size of the data fork, not the total size of data fork
1387 plus resource fork. (S<Mac OS>).
1389 C<-s> by name on an open file will return the space reserved on disk,
1390 rather than the current extent. C<-s> on an open filehandle returns the
1391 current size. (S<RISC OS>)
1393 C<-R>, C<-W>, C<-X>, C<-O> are indistinguishable from C<-r>, C<-w>,
1394 C<-x>, C<-o>. (S<Mac OS>, Win32, VMS, S<RISC OS>)
1396 C<-b>, C<-c>, C<-k>, C<-g>, C<-p>, C<-u>, C<-A> are not implemented.
1399 C<-g>, C<-k>, C<-l>, C<-p>, C<-u>, C<-A> are not particularly meaningful.
1400 (Win32, VMS, S<RISC OS>)
1402 C<-d> is true if passed a device spec without an explicit directory.
1405 C<-T> and C<-B> are implemented, but might misclassify Mac text files
1406 with foreign characters; this is the case will all platforms, but may
1407 affect S<Mac OS> often. (S<Mac OS>)
1409 C<-x> (or C<-X>) determine if a file ends in one of the executable
1410 suffixes. C<-S> is meaningless. (Win32)
1412 C<-x> (or C<-X>) determine if a file has an executable file type.
1419 Not implemented. (Win32)
1421 =item binmode FILEHANDLE
1423 Meaningless. (S<Mac OS>, S<RISC OS>)
1425 Reopens file and restores pointer; if function fails, underlying
1426 filehandle may be closed, or pointer may be in a different position.
1429 The value returned by C<tell> may be affected after the call, and
1430 the filehandle may be flushed. (Win32)
1434 Only limited meaning. Disabling/enabling write permission is mapped to
1435 locking/unlocking the file. (S<Mac OS>)
1437 Only good for changing "owner" read-write access, "group", and "other"
1438 bits are meaningless. (Win32)
1440 Only good for changing "owner" and "other" read-write access. (S<RISC OS>)
1442 Access permissions are mapped onto VOS access-control list changes. (VOS)
1444 The actual permissions set depend on the value of the C<CYGWIN>
1445 in the SYSTEM environment settings. (Cygwin)
1449 Not implemented. (S<Mac OS>, Win32, Plan9, S<RISC OS>, VOS)
1451 Does nothing, but won't fail. (Win32)
1453 =item chroot FILENAME
1457 Not implemented. (S<Mac OS>, Win32, VMS, Plan9, S<RISC OS>, VOS, VM/ESA)
1459 =item crypt PLAINTEXT,SALT
1461 May not be available if library or source was not provided when building
1464 Not implemented. (VOS)
1468 Not implemented. (VMS, Plan9, VOS)
1470 =item dbmopen HASH,DBNAME,MODE
1472 Not implemented. (VMS, Plan9, VOS)
1476 Not useful. (S<Mac OS>, S<RISC OS>)
1478 Not implemented. (Win32)
1480 Invokes VMS debugger. (VMS)
1484 Not implemented. (S<Mac OS>)
1486 Implemented via Spawn. (VM/ESA)
1488 Does not automatically flush output handles on some platforms.
1489 (SunOS, Solaris, HP-UX)
1495 Emulates UNIX exit() (which considers C<exit 1> to indicate an error) by
1496 mapping the C<1> to SS$_ABORT (C<44>). This behavior may be overridden
1497 with the pragma C<use vmsish 'exit'>. As with the CRTL's exit()
1498 function, C<exit 0> is also mapped to an exit status of SS$_NORMAL
1499 (C<1>); this mapping cannot be overridden. Any other argument to exit()
1500 is used directly as Perl's exit status. (VMS)
1502 =item fcntl FILEHANDLE,FUNCTION,SCALAR
1504 Not implemented. (Win32, VMS)
1506 =item flock FILEHANDLE,OPERATION
1508 Not implemented (S<Mac OS>, VMS, S<RISC OS>, VOS).
1510 Available only on Windows NT (not on Windows 95). (Win32)
1514 Not implemented. (S<Mac OS>, AmigaOS, S<RISC OS>, VOS, VM/ESA)
1516 Emulated using multiple interpreters. See L<perlfork>. (Win32)
1518 Does not automatically flush output handles on some platforms.
1519 (SunOS, Solaris, HP-UX)
1523 Not implemented. (S<Mac OS>, S<RISC OS>)
1527 Not implemented. (S<Mac OS>, Win32, VMS, S<RISC OS>, VOS)
1531 Not implemented. (S<Mac OS>, Win32, VMS, S<RISC OS>)
1533 =item getpriority WHICH,WHO
1535 Not implemented. (S<Mac OS>, Win32, VMS, S<RISC OS>, VOS, VM/ESA)
1539 Not implemented. (S<Mac OS>, Win32)
1541 Not useful. (S<RISC OS>)
1545 Not implemented. (S<Mac OS>, Win32, VMS, S<RISC OS>)
1547 =item getnetbyname NAME
1549 Not implemented. (S<Mac OS>, Win32, Plan9)
1553 Not implemented. (S<Mac OS>, Win32)
1555 Not useful. (S<RISC OS>)
1559 Not implemented. (S<Mac OS>, Win32, VMS, S<RISC OS>)
1561 =item getnetbyaddr ADDR,ADDRTYPE
1563 Not implemented. (S<Mac OS>, Win32, Plan9)
1565 =item getprotobynumber NUMBER
1567 Not implemented. (S<Mac OS>)
1569 =item getservbyport PORT,PROTO
1571 Not implemented. (S<Mac OS>)
1575 Not implemented. (S<Mac OS>, Win32, VM/ESA)
1579 Not implemented. (S<Mac OS>, Win32, VMS, VM/ESA)
1583 Not implemented. (S<Mac OS>, Win32)
1587 Not implemented. (S<Mac OS>, Win32, Plan9)
1591 Not implemented. (S<Mac OS>, Win32, Plan9)
1595 Not implemented. (Win32, Plan9)
1599 Not implemented. (S<Mac OS>, Win32, S<RISC OS>)
1603 Not implemented. (S<Mac OS>, Win32, VMS, S<RISC OS>)
1605 =item sethostent STAYOPEN
1607 Not implemented. (S<Mac OS>, Win32, Plan9, S<RISC OS>)
1609 =item setnetent STAYOPEN
1611 Not implemented. (S<Mac OS>, Win32, Plan9, S<RISC OS>)
1613 =item setprotoent STAYOPEN
1615 Not implemented. (S<Mac OS>, Win32, Plan9, S<RISC OS>)
1617 =item setservent STAYOPEN
1619 Not implemented. (Plan9, Win32, S<RISC OS>)
1623 Not implemented. (S<Mac OS>, MPE/iX, VM/ESA, Win32)
1627 Not implemented. (S<Mac OS>, MPE/iX, S<RISC OS>, VM/ESA, VMS, Win32)
1631 Not implemented. (S<Mac OS>, Win32)
1635 Not implemented. (S<Mac OS>, Win32, Plan9)
1639 Not implemented. (S<Mac OS>, Win32, Plan9)
1643 Not implemented. (Plan9, Win32)
1645 =item getsockopt SOCKET,LEVEL,OPTNAME
1647 Not implemented. (Plan9)
1653 This operator is implemented via the File::Glob extension on most
1654 platforms. See L<File::Glob> for portability information.
1656 =item ioctl FILEHANDLE,FUNCTION,SCALAR
1658 Not implemented. (VMS)
1660 Available only for socket handles, and it does what the ioctlsocket() call
1661 in the Winsock API does. (Win32)
1663 Available only for socket handles. (S<RISC OS>)
1665 =item kill SIGNAL, LIST
1667 C<kill(0, LIST)> is implemented for the sake of taint checking;
1668 use with other signals is unimplemented. (S<Mac OS>)
1670 Not implemented, hence not useful for taint checking. (S<RISC OS>)
1672 C<kill()> doesn't have the semantics of C<raise()>, i.e. it doesn't send
1673 a signal to the identified process like it does on Unix platforms.
1674 Instead C<kill($sig, $pid)> terminates the process identified by $pid,
1675 and makes it exit immediately with exit status $sig. As in Unix, if
1676 $sig is 0 and the specified process exists, it returns true without
1677 actually terminating it. (Win32)
1679 =item link OLDFILE,NEWFILE
1681 Not implemented. (S<Mac OS>, MPE/iX, VMS, S<RISC OS>)
1683 Link count not updated because hard links are not quite that hard
1684 (They are sort of half-way between hard and soft links). (AmigaOS)
1686 Hard links are implemented on Win32 (Windows NT and Windows 2000)
1689 =item lstat FILEHANDLE
1695 Not implemented. (VMS, S<RISC OS>)
1697 Return values (especially for device and inode) may be bogus. (Win32)
1699 =item msgctl ID,CMD,ARG
1701 =item msgget KEY,FLAGS
1703 =item msgsnd ID,MSG,FLAGS
1705 =item msgrcv ID,VAR,SIZE,TYPE,FLAGS
1707 Not implemented. (S<Mac OS>, Win32, VMS, Plan9, S<RISC OS>, VOS)
1709 =item open FILEHANDLE,EXPR
1711 =item open FILEHANDLE
1713 The C<|> variants are supported only if ToolServer is installed.
1716 open to C<|-> and C<-|> are unsupported. (S<Mac OS>, Win32, S<RISC OS>)
1718 Opening a process does not automatically flush output handles on some
1719 platforms. (SunOS, Solaris, HP-UX)
1721 =item pipe READHANDLE,WRITEHANDLE
1723 Very limited functionality. (MiNT)
1729 Not implemented. (Win32, VMS, S<RISC OS>)
1731 =item select RBITS,WBITS,EBITS,TIMEOUT
1733 Only implemented on sockets. (Win32, VMS)
1735 Only reliable on sockets. (S<RISC OS>)
1737 Note that the C<select FILEHANDLE> form is generally portable.
1739 =item semctl ID,SEMNUM,CMD,ARG
1741 =item semget KEY,NSEMS,FLAGS
1743 =item semop KEY,OPSTRING
1745 Not implemented. (S<Mac OS>, Win32, VMS, S<RISC OS>, VOS)
1749 Not implemented. (MPE/iX, Win32)
1751 =item setpgrp PID,PGRP
1753 Not implemented. (S<Mac OS>, Win32, VMS, S<RISC OS>, VOS)
1755 =item setpriority WHICH,WHO,PRIORITY
1757 Not implemented. (S<Mac OS>, Win32, VMS, S<RISC OS>, VOS)
1761 Not implemented. (MPE/iX, Win32)
1763 =item setsockopt SOCKET,LEVEL,OPTNAME,OPTVAL
1765 Not implemented. (Plan9)
1767 =item shmctl ID,CMD,ARG
1769 =item shmget KEY,SIZE,FLAGS
1771 =item shmread ID,VAR,POS,SIZE
1773 =item shmwrite ID,STRING,POS,SIZE
1775 Not implemented. (S<Mac OS>, Win32, VMS, S<RISC OS>, VOS)
1777 =item sockatmark SOCKET
1779 A relatively recent addition to socket functions, may not
1780 be implemented even in UNIX platforms.
1782 =item socketpair SOCKET1,SOCKET2,DOMAIN,TYPE,PROTOCOL
1784 Not implemented. (Win32, VMS, S<RISC OS>, VOS, VM/ESA)
1786 =item stat FILEHANDLE
1792 Platforms that do not have rdev, blksize, or blocks will return these
1793 as '', so numeric comparison or manipulation of these fields may cause
1794 'not numeric' warnings.
1796 mtime and atime are the same thing, and ctime is creation time instead of
1797 inode change time. (S<Mac OS>).
1799 ctime not supported on UFS (S<Mac OS X>).
1801 ctime is creation time instead of inode change time (Win32).
1803 device and inode are not meaningful. (Win32)
1805 device and inode are not necessarily reliable. (VMS)
1807 mtime, atime and ctime all return the last modification time. Device and
1808 inode are not necessarily reliable. (S<RISC OS>)
1810 dev, rdev, blksize, and blocks are not available. inode is not
1811 meaningful and will differ between stat calls on the same file. (os2)
1813 some versions of cygwin when doing a stat("foo") and if not finding it
1814 may then attempt to stat("foo.exe") (Cygwin)
1816 =item symlink OLDFILE,NEWFILE
1818 Not implemented. (Win32, VMS, S<RISC OS>)
1822 Not implemented. (S<Mac OS>, Win32, VMS, S<RISC OS>, VOS, VM/ESA)
1824 =item sysopen FILEHANDLE,FILENAME,MODE,PERMS
1826 The traditional "0", "1", and "2" MODEs are implemented with different
1827 numeric values on some systems. The flags exported by C<Fcntl>
1828 (O_RDONLY, O_WRONLY, O_RDWR) should work everywhere though. (S<Mac
1829 OS>, OS/390, VM/ESA)
1833 In general, do not assume the UNIX/POSIX semantics that you can shift
1834 C<$?> right by eight to get the exit value, or that C<$? & 127>
1835 would give you the number of the signal that terminated the program,
1836 or that C<$? & 128> would test true if the program was terminated by a
1837 coredump. Instead, use the POSIX W*() interfaces: for example, use
1838 WIFEXITED($?) an WEXITVALUE($?) to test for a normal exit and the exit
1839 value, and WIFSIGNALED($?) and WTERMSIG($?) for a signal exit and the
1840 signal. Core dumping is not a portable concept, so there's no portable
1841 way to test for that.
1843 Only implemented if ToolServer is installed. (S<Mac OS>)
1845 As an optimization, may not call the command shell specified in
1846 C<$ENV{PERL5SHELL}>. C<system(1, @args)> spawns an external
1847 process and immediately returns its process designator, without
1848 waiting for it to terminate. Return value may be used subsequently
1849 in C<wait> or C<waitpid>. Failure to spawn() a subprocess is indicated
1850 by setting $? to "255 << 8". C<$?> is set in a way compatible with
1851 Unix (i.e. the exitstatus of the subprocess is obtained by "$? >> 8",
1852 as described in the documentation). (Win32)
1854 There is no shell to process metacharacters, and the native standard is
1855 to pass a command line terminated by "\n" "\r" or "\0" to the spawned
1856 program. Redirection such as C<< > foo >> is performed (if at all) by
1857 the run time library of the spawned program. C<system> I<list> will call
1858 the Unix emulation library's C<exec> emulation, which attempts to provide
1859 emulation of the stdin, stdout, stderr in force in the parent, providing
1860 the child program uses a compatible version of the emulation library.
1861 I<scalar> will call the native command line direct and no such emulation
1862 of a child Unix program will exists. Mileage B<will> vary. (S<RISC OS>)
1864 Far from being POSIX compliant. Because there may be no underlying
1865 /bin/sh tries to work around the problem by forking and execing the
1866 first token in its argument string. Handles basic redirection
1867 ("<" or ">") on its own behalf. (MiNT)
1869 Does not automatically flush output handles on some platforms.
1870 (SunOS, Solaris, HP-UX)
1872 The return value is POSIX-like (shifted up by 8 bits), which only allows
1873 room for a made-up value derived from the severity bits of the native
1874 32-bit condition code (unless overridden by C<use vmsish 'status'>).
1875 For more details see L<perlvms/$?>. (VMS)
1879 Only the first entry returned is nonzero. (S<Mac OS>)
1881 "cumulative" times will be bogus. On anything other than Windows NT
1882 or Windows 2000, "system" time will be bogus, and "user" time is
1883 actually the time returned by the clock() function in the C runtime
1886 Not useful. (S<RISC OS>)
1888 =item truncate FILEHANDLE,LENGTH
1890 =item truncate EXPR,LENGTH
1892 Not implemented. (Older versions of VMS)
1894 Truncation to zero-length only. (VOS)
1896 If a FILEHANDLE is supplied, it must be writable and opened in append
1897 mode (i.e., use C<<< open(FH, '>>filename') >>>
1898 or C<sysopen(FH,...,O_APPEND|O_RDWR)>. If a filename is supplied, it
1899 should not be held open elsewhere. (Win32)
1905 Returns undef where unavailable, as of version 5.005.
1907 C<umask> works but the correct permissions are set only when the file
1908 is finally closed. (AmigaOS)
1912 Only the modification time is updated. (S<BeOS>, S<Mac OS>, VMS, S<RISC OS>)
1914 May not behave as expected. Behavior depends on the C runtime
1915 library's implementation of utime(), and the filesystem being
1916 used. The FAT filesystem typically does not support an "access
1917 time" field, and it may limit timestamps to a granularity of
1918 two seconds. (Win32)
1922 =item waitpid PID,FLAGS
1924 Not implemented. (S<Mac OS>, VOS)
1926 Can only be applied to process handles returned for processes spawned
1927 using C<system(1, ...)> or pseudo processes created with C<fork()>. (Win32)
1929 Not useful. (S<RISC OS>)
1937 =item v1.48, 02 February 2001
1939 Various updates from perl5-porters over the past year, supported
1940 platforms update from Jarkko Hietaniemi.
1942 =item v1.47, 22 March 2000
1944 Various cleanups from Tom Christiansen, including migration of
1945 long platform listings from L<perl>.
1947 =item v1.46, 12 February 2000
1949 Updates for VOS and MPE/iX. (Peter Prymmer) Other small changes.
1951 =item v1.45, 20 December 1999
1953 Small changes from 5.005_63 distribution, more changes to EBCDIC info.
1955 =item v1.44, 19 July 1999
1957 A bunch of updates from Peter Prymmer for C<$^O> values,
1958 endianness, File::Spec, VMS, BS2000, OS/400.
1960 =item v1.43, 24 May 1999
1962 Added a lot of cleaning up from Tom Christiansen.
1964 =item v1.42, 22 May 1999
1966 Added notes about tests, sprintf/printf, and epoch offsets.
1968 =item v1.41, 19 May 1999
1970 Lots more little changes to formatting and content.
1972 Added a bunch of C<$^O> and related values
1973 for various platforms; fixed mail and web addresses, and added
1974 and changed miscellaneous notes. (Peter Prymmer)
1976 =item v1.40, 11 April 1999
1978 Miscellaneous changes.
1980 =item v1.39, 11 February 1999
1982 Changes from Jarkko and EMX URL fixes Michael Schwern. Additional
1983 note about newlines added.
1985 =item v1.38, 31 December 1998
1987 More changes from Jarkko.
1989 =item v1.37, 19 December 1998
1991 More minor changes. Merge two separate version 1.35 documents.
1993 =item v1.36, 9 September 1998
1995 Updated for Stratus VOS. Also known as version 1.35.
1997 =item v1.35, 13 August 1998
1999 Integrate more minor changes, plus addition of new sections under
2000 L<"ISSUES">: L<"Numbers endianness and Width">,
2001 L<"Character sets and character encoding">,
2002 L<"Internationalisation">.
2004 =item v1.33, 06 August 1998
2006 Integrate more minor changes.
2008 =item v1.32, 05 August 1998
2010 Integrate more minor changes.
2012 =item v1.30, 03 August 1998
2014 Major update for RISC OS, other minor changes.
2016 =item v1.23, 10 July 1998
2018 First public release with perl5.005.
2022 =head1 Supported Platforms
2024 As of early 2001 (the Perl releases 5.6.1 and 5.7.1), the following
2025 platforms are able to build Perl from the standard source code
2026 distribution available at http://www.cpan.org/src/index.html
2049 Tru64 UNIX (DEC OSF/1, Digital UNIX)
2055 1) in DOS mode either the DOS or OS/2 ports can be used
2056 2) Mac OS Classic (pre-X) is almost 5.6.1-ready; building from
2057 the source does work with 5.6.1, but additional MacOS specific
2058 source code is needed for a complete build. See the web
2059 site http://dev.macperl.org/ for more information.
2060 3) compilers: Borland, Cygwin, Mingw32 EGCS/GCC, VC++
2062 The following platforms worked for the previous releases (5.6.0 and 5.7.0),
2063 but we did not manage to test these in time for the 5.7.1 release.
2064 There is a very good chance that these will work fine with the 5.7.1.
2082 The following platform worked for the 5.005_03 major release but not
2083 for 5.6.0. Standardization on UTF-8 as the internal string
2084 representation in 5.6.0 and 5.6.1 introduced incompatibilities in this
2085 EBCDIC platform. While Perl 5.7.1 will build on this platform some
2086 regression tests may fail and the C<use utf8;> pragma typically
2087 introduces text handling errors.
2091 1) previously known as MVS, about to become z/OS.
2093 Strongly related to the OS/390 platform by also being EBCDIC-based
2094 mainframe platforms are the following platforms:
2099 These are also expected to work, albeit with no UTF-8 support, under 5.6.1
2100 for the same reasons as OS/390. Contact the mailing list perl-mvs@perl.org
2103 The following platforms have been known to build Perl from source in
2104 the past (5.005_03 and earlier), but we haven't been able to verify
2105 their status for the current release, either because the
2106 hardware/software platforms are rare or because we don't have an
2107 active champion on these platforms--or both. They used to work,
2108 though, so go ahead and try compiling them, and let perlbug@perl.org
2147 Support for the following platform is planned for a future Perl release:
2151 The following platforms have their own source code distributions and
2152 binaries available via http://www.cpan.org/ports/
2158 Tandem Guardian 5.004
2160 The following platforms have only binaries available via
2161 http://www.cpan.org/ports/index.html :
2165 Acorn RISCOS 5.005_02
2169 Although we do suggest that you always build your own Perl from
2170 the source code, both for maximal configurability and for security,
2171 in case you are in a hurry you can check
2172 http://www.cpan.org/ports/index.html for binary distributions.
2176 L<perlaix>, L<perlapollo>, L<perlamiga>, L<perlbeos>, L<perlbs200>,
2177 L<perlce>, L<perlcygwin>, L<perldgux>, L<perldos>, L<perlepoc>, L<perlebcdic>,
2178 L<perlhurd>, L<perlhpux>, L<perlmachten>, L<perlmacos>, L<perlmint>,
2179 L<perlmpeix>, L<perlnetware>, L<perlos2>, L<perlos390>, L<perlplan9>,
2180 L<perlqnx>, L<perlsolaris>, L<perltru64>, L<perlunicode>,
2181 L<perlvmesa>, L<perlvms>, L<perlvos>, L<perlwin32>, and L<Win32>.
2183 =head1 AUTHORS / CONTRIBUTORS
2185 Abigail <abigail@foad.org>,
2186 Charles Bailey <bailey@newman.upenn.edu>,
2187 Graham Barr <gbarr@pobox.com>,
2188 Tom Christiansen <tchrist@perl.com>,
2189 Nicholas Clark <nick@ccl4.org>,
2190 Thomas Dorner <Thomas.Dorner@start.de>,
2191 Andy Dougherty <doughera@lafayette.edu>,
2192 Dominic Dunlop <domo@computer.org>,
2193 Neale Ferguson <neale@vma.tabnsw.com.au>,
2194 David J. Fiander <davidf@mks.com>,
2195 Paul Green <Paul_Green@stratus.com>,
2196 M.J.T. Guy <mjtg@cam.ac.uk>,
2197 Jarkko Hietaniemi <jhi@iki.fi>,
2198 Luther Huffman <lutherh@stratcom.com>,
2199 Nick Ing-Simmons <nick@ing-simmons.net>,
2200 Andreas J. KE<ouml>nig <a.koenig@mind.de>,
2201 Markus Laker <mlaker@contax.co.uk>,
2202 Andrew M. Langmead <aml@world.std.com>,
2203 Larry Moore <ljmoore@freespace.net>,
2204 Paul Moore <Paul.Moore@uk.origin-it.com>,
2205 Chris Nandor <pudge@pobox.com>,
2206 Matthias Neeracher <neeracher@mac.com>,
2207 Philip Newton <pne@cpan.org>,
2208 Gary Ng <71564.1743@CompuServe.COM>,
2209 Tom Phoenix <rootbeer@teleport.com>,
2210 AndrE<eacute> Pirard <A.Pirard@ulg.ac.be>,
2211 Peter Prymmer <pvhp@forte.com>,
2212 Hugo van der Sanden <hv@crypt0.demon.co.uk>,
2213 Gurusamy Sarathy <gsar@activestate.com>,
2214 Paul J. Schinder <schinder@pobox.com>,
2215 Michael G Schwern <schwern@pobox.com>,
2216 Dan Sugalski <dan@sidhe.org>,
2217 Nathan Torkington <gnat@frii.com>.