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 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 $ perl -ne "print if /perl_setup/i" >system>notices
1028 $ perl -ne "print if /perl_setup/i" /system/notices
1030 or even a mixture of both as in:
1032 $ 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 See F<README.vos> for restrictions that apply when Perl is built
1042 with the alpha version of VOS POSIX.1 support.
1044 Perl on VOS is built without any extensions and does not support
1047 The value of C<$^O> on VOS is "VOS". To determine the architecture that
1048 you are running on without resorting to loading all of C<%Config> you
1049 can examine the content of the @INC array like so:
1052 print "I'm on a Stratus box!\n";
1054 print "I'm not on a Stratus box!\n";
1058 if (grep(/860/, @INC)) {
1059 print "This box is a Stratus XA/R!\n";
1061 } elsif (grep(/7100/, @INC)) {
1062 print "This box is a Stratus HP 7100 or 8xxx!\n";
1064 } elsif (grep(/8000/, @INC)) {
1065 print "This box is a Stratus HP 8xxx!\n";
1068 print "This box is a Stratus 68K!\n";
1081 The VOS mailing list.
1083 There is no specific mailing list for Perl on VOS. You can post
1084 comments to the comp.sys.stratus newsgroup, or subscribe to the general
1085 Stratus mailing list. Send a letter with "Subscribe Info-Stratus" in
1086 the message body to majordomo@list.stratagy.com.
1090 VOS Perl on the web at http://ftp.stratus.com/pub/vos/vos.html
1094 =head2 EBCDIC Platforms
1096 Recent versions of Perl have been ported to platforms such as OS/400 on
1097 AS/400 minicomputers as well as OS/390, VM/ESA, and BS2000 for S/390
1098 Mainframes. Such computers use EBCDIC character sets internally (usually
1099 Character Code Set ID 0037 for OS/400 and either 1047 or POSIX-BC for S/390
1100 systems). On the mainframe perl currently works under the "Unix system
1101 services for OS/390" (formerly known as OpenEdition), VM/ESA OpenEdition, or
1102 the BS200 POSIX-BC system (BS2000 is supported in perl 5.6 and greater).
1103 See L<perlos390> for details.
1105 As of R2.5 of USS for OS/390 and Version 2.3 of VM/ESA these Unix
1106 sub-systems do not support the C<#!> shebang trick for script invocation.
1107 Hence, on OS/390 and VM/ESA perl scripts can be executed with a header
1108 similar to the following simple script:
1111 eval 'exec /usr/local/bin/perl -S $0 ${1+"$@"}'
1113 #!/usr/local/bin/perl # just a comment really
1115 print "Hello from perl!\n";
1117 OS/390 will support the C<#!> shebang trick in release 2.8 and beyond.
1118 Calls to C<system> and backticks can use POSIX shell syntax on all
1121 On the AS/400, if PERL5 is in your library list, you may need
1122 to wrap your perl scripts in a CL procedure to invoke them like so:
1125 CALL PGM(PERL5/PERL) PARM('/QOpenSys/hello.pl')
1128 This will invoke the perl script F<hello.pl> in the root of the
1129 QOpenSys file system. On the AS/400 calls to C<system> or backticks
1132 On these platforms, bear in mind that the EBCDIC character set may have
1133 an effect on what happens with some perl functions (such as C<chr>,
1134 C<pack>, C<print>, C<printf>, C<ord>, C<sort>, C<sprintf>, C<unpack>), as
1135 well as bit-fiddling with ASCII constants using operators like C<^>, C<&>
1136 and C<|>, not to mention dealing with socket interfaces to ASCII computers
1137 (see L<"Newlines">).
1139 Fortunately, most web servers for the mainframe will correctly
1140 translate the C<\n> in the following statement to its ASCII equivalent
1141 (C<\r> is the same under both Unix and OS/390 & VM/ESA):
1143 print "Content-type: text/html\r\n\r\n";
1145 The values of C<$^O> on some of these platforms includes:
1147 uname $^O $Config{'archname'}
1148 --------------------------------------------
1151 POSIX-BC posix-bc BS2000-posix-bc
1154 Some simple tricks for determining if you are running on an EBCDIC
1155 platform could include any of the following (perhaps all):
1157 if ("\t" eq "\05") { print "EBCDIC may be spoken here!\n"; }
1159 if (ord('A') == 193) { print "EBCDIC may be spoken here!\n"; }
1161 if (chr(169) eq 'z') { print "EBCDIC may be spoken here!\n"; }
1163 One thing you may not want to rely on is the EBCDIC encoding
1164 of punctuation characters since these may differ from code page to code
1165 page (and once your module or script is rumoured to work with EBCDIC,
1166 folks will want it to work with all EBCDIC character sets).
1176 L<perlos390>, F<README.os390>, F<perlbs2000>, F<README.vmesa>,
1181 The perl-mvs@perl.org list is for discussion of porting issues as well as
1182 general usage issues for all EBCDIC Perls. Send a message body of
1183 "subscribe perl-mvs" to majordomo@perl.org.
1187 AS/400 Perl information at
1188 http://as400.rochester.ibm.com/
1189 as well as on CPAN in the F<ports/> directory.
1193 =head2 Acorn RISC OS
1195 Because Acorns use ASCII with newlines (C<\n>) in text files as C<\012> like
1196 Unix, and because Unix filename emulation is turned on by default,
1197 most simple scripts will probably work "out of the box". The native
1198 filesystem is modular, and individual filesystems are free to be
1199 case-sensitive or insensitive, and are usually case-preserving. Some
1200 native filesystems have name length limits, which file and directory
1201 names are silently truncated to fit. Scripts should be aware that the
1202 standard filesystem currently has a name length limit of B<10>
1203 characters, with up to 77 items in a directory, but other filesystems
1204 may not impose such limitations.
1206 Native filenames are of the form
1208 Filesystem#Special_Field::DiskName.$.Directory.Directory.File
1212 Special_Field is not usually present, but may contain . and $ .
1213 Filesystem =~ m|[A-Za-z0-9_]|
1214 DsicName =~ m|[A-Za-z0-9_/]|
1215 $ represents the root directory
1216 . is the path separator
1217 @ is the current directory (per filesystem but machine global)
1218 ^ is the parent directory
1219 Directory and File =~ m|[^\0- "\.\$\%\&:\@\\^\|\177]+|
1221 The default filename translation is roughly C<tr|/.|./|;>
1223 Note that C<"ADFS::HardDisk.$.File" ne 'ADFS::HardDisk.$.File'> and that
1224 the second stage of C<$> interpolation in regular expressions will fall
1225 foul of the C<$.> if scripts are not careful.
1227 Logical paths specified by system variables containing comma-separated
1228 search lists are also allowed; hence C<System:Modules> is a valid
1229 filename, and the filesystem will prefix C<Modules> with each section of
1230 C<System$Path> until a name is made that points to an object on disk.
1231 Writing to a new file C<System:Modules> would be allowed only if
1232 C<System$Path> contains a single item list. The filesystem will also
1233 expand system variables in filenames if enclosed in angle brackets, so
1234 C<< <System$Dir>.Modules >> would look for the file
1235 S<C<$ENV{'System$Dir'} . 'Modules'>>. The obvious implication of this is
1236 that B<fully qualified filenames can start with C<< <> >>> and should
1237 be protected when C<open> is used for input.
1239 Because C<.> was in use as a directory separator and filenames could not
1240 be assumed to be unique after 10 characters, Acorn implemented the C
1241 compiler to strip the trailing C<.c> C<.h> C<.s> and C<.o> suffix from
1242 filenames specified in source code and store the respective files in
1243 subdirectories named after the suffix. Hence files are translated:
1246 C:foo.h C:h.foo (logical path variable)
1247 sys/os.h sys.h.os (C compiler groks Unix-speak)
1248 10charname.c c.10charname
1249 10charname.o o.10charname
1250 11charname_.c c.11charname (assuming filesystem truncates at 10)
1252 The Unix emulation library's translation of filenames to native assumes
1253 that this sort of translation is required, and it allows a user-defined list
1254 of known suffixes that it will transpose in this fashion. This may
1255 seem transparent, but consider that with these rules C<foo/bar/baz.h>
1256 and C<foo/bar/h/baz> both map to C<foo.bar.h.baz>, and that C<readdir> and
1257 C<glob> cannot and do not attempt to emulate the reverse mapping. Other
1258 C<.>'s in filenames are translated to C</>.
1260 As implied above, the environment accessed through C<%ENV> is global, and
1261 the convention is that program specific environment variables are of the
1262 form C<Program$Name>. Each filesystem maintains a current directory,
1263 and the current filesystem's current directory is the B<global> current
1264 directory. Consequently, sociable programs don't change the current
1265 directory but rely on full pathnames, and programs (and Makefiles) cannot
1266 assume that they can spawn a child process which can change the current
1267 directory without affecting its parent (and everyone else for that
1270 Because native operating system filehandles are global and are currently
1271 allocated down from 255, with 0 being a reserved value, the Unix emulation
1272 library emulates Unix filehandles. Consequently, you can't rely on
1273 passing C<STDIN>, C<STDOUT>, or C<STDERR> to your children.
1275 The desire of users to express filenames of the form
1276 C<< <Foo$Dir>.Bar >> on the command line unquoted causes problems,
1277 too: C<``> command output capture has to perform a guessing game. It
1278 assumes that a string C<< <[^<>]+\$[^<>]> >> is a
1279 reference to an environment variable, whereas anything else involving
1280 C<< < >> or C<< > >> is redirection, and generally manages to be 99%
1281 right. Of course, the problem remains that scripts cannot rely on any
1282 Unix tools being available, or that any tools found have Unix-like command
1285 Extensions and XS are, in theory, buildable by anyone using free
1286 tools. In practice, many don't, as users of the Acorn platform are
1287 used to binary distributions. MakeMaker does run, but no available
1288 make currently copes with MakeMaker's makefiles; even if and when
1289 this should be fixed, the lack of a Unix-like shell will cause
1290 problems with makefile rules, especially lines of the form C<cd
1291 sdbm && make all>, and anything using quoting.
1293 "S<RISC OS>" is the proper name for the operating system, but the value
1294 in C<$^O> is "riscos" (because we don't like shouting).
1298 Perl has been ported to many platforms that do not fit into any of
1299 the categories listed above. Some, such as AmigaOS, Atari MiNT,
1300 BeOS, HP MPE/iX, QNX, Plan 9, and VOS, have been well-integrated
1301 into the standard Perl source code kit. You may need to see the
1302 F<ports/> directory on CPAN for information, and possibly binaries,
1303 for the likes of: aos, Atari ST, lynxos, riscos, Novell Netware,
1304 Tandem Guardian, I<etc.> (Yes, we know that some of these OSes may
1305 fall under the Unix category, but we are not a standards body.)
1307 Some approximate operating system names and their C<$^O> values
1308 in the "OTHER" category include:
1310 OS $^O $Config{'archname'}
1311 ------------------------------------------
1312 Amiga DOS amigaos m68k-amigos
1313 MPE/iX mpeix PA-RISC1.1
1321 Amiga, F<README.amiga> (installed as L<perlamiga>).
1325 Atari, F<README.mint> and Guido Flohr's web page
1326 http://stud.uni-sb.de/~gufl0000/
1330 Be OS, F<README.beos>
1334 HP 300 MPE/iX, F<README.mpeix> and Mark Bixby's web page
1335 http://www.bixby.org/mark/perlix.html
1339 A free perl5-based PERL.NLM for Novell Netware is available in
1340 precompiled binary and source code form from http://www.novell.com/
1341 as well as from CPAN.
1345 Plan 9, F<README.plan9>
1349 =head1 FUNCTION IMPLEMENTATIONS
1351 Listed below are functions that are either completely unimplemented
1352 or else have been implemented differently on various platforms.
1353 Following each description will be, in parentheses, a list of
1354 platforms that the description applies to.
1356 The list may well be incomplete, or even wrong in some places. When
1357 in doubt, consult the platform-specific README files in the Perl
1358 source distribution, and any other documentation resources accompanying
1361 Be aware, moreover, that even among Unix-ish systems there are variations.
1363 For many functions, you can also query C<%Config>, exported by
1364 default from the Config module. For example, to check whether the
1365 platform has the C<lstat> call, check C<$Config{d_lstat}>. See
1366 L<Config> for a full description of available variables.
1368 =head2 Alphabetical Listing of Perl Functions
1378 C<-r>, C<-w>, and C<-x> have a limited meaning only; directories
1379 and applications are executable, and there are no uid/gid
1380 considerations. C<-o> is not supported. (S<Mac OS>)
1382 C<-r>, C<-w>, C<-x>, and C<-o> tell whether the file is accessible,
1383 which may not reflect UIC-based file protections. (VMS)
1385 C<-s> returns the size of the data fork, not the total size of data fork
1386 plus resource fork. (S<Mac OS>).
1388 C<-s> by name on an open file will return the space reserved on disk,
1389 rather than the current extent. C<-s> on an open filehandle returns the
1390 current size. (S<RISC OS>)
1392 C<-R>, C<-W>, C<-X>, C<-O> are indistinguishable from C<-r>, C<-w>,
1393 C<-x>, C<-o>. (S<Mac OS>, Win32, VMS, S<RISC OS>)
1395 C<-b>, C<-c>, C<-k>, C<-g>, C<-p>, C<-u>, C<-A> are not implemented.
1398 C<-g>, C<-k>, C<-l>, C<-p>, C<-u>, C<-A> are not particularly meaningful.
1399 (Win32, VMS, S<RISC OS>)
1401 C<-d> is true if passed a device spec without an explicit directory.
1404 C<-T> and C<-B> are implemented, but might misclassify Mac text files
1405 with foreign characters; this is the case will all platforms, but may
1406 affect S<Mac OS> often. (S<Mac OS>)
1408 C<-x> (or C<-X>) determine if a file ends in one of the executable
1409 suffixes. C<-S> is meaningless. (Win32)
1411 C<-x> (or C<-X>) determine if a file has an executable file type.
1418 Not implemented. (Win32)
1420 =item binmode FILEHANDLE
1422 Meaningless. (S<Mac OS>, S<RISC OS>)
1424 Reopens file and restores pointer; if function fails, underlying
1425 filehandle may be closed, or pointer may be in a different position.
1428 The value returned by C<tell> may be affected after the call, and
1429 the filehandle may be flushed. (Win32)
1433 Only limited meaning. Disabling/enabling write permission is mapped to
1434 locking/unlocking the file. (S<Mac OS>)
1436 Only good for changing "owner" read-write access, "group", and "other"
1437 bits are meaningless. (Win32)
1439 Only good for changing "owner" and "other" read-write access. (S<RISC OS>)
1441 Access permissions are mapped onto VOS access-control list changes. (VOS)
1443 The actual permissions set depend on the value of the C<CYGWIN>
1444 in the SYSTEM environment settings. (Cygwin)
1448 Not implemented. (S<Mac OS>, Win32, Plan9, S<RISC OS>, VOS)
1450 Does nothing, but won't fail. (Win32)
1452 =item chroot FILENAME
1456 Not implemented. (S<Mac OS>, Win32, VMS, Plan9, S<RISC OS>, VOS, VM/ESA)
1458 =item crypt PLAINTEXT,SALT
1460 May not be available if library or source was not provided when building
1463 Not implemented. (VOS)
1467 Not implemented. (VMS, Plan9, VOS)
1469 =item dbmopen HASH,DBNAME,MODE
1471 Not implemented. (VMS, Plan9, VOS)
1475 Not useful. (S<Mac OS>, S<RISC OS>)
1477 Not implemented. (Win32)
1479 Invokes VMS debugger. (VMS)
1483 Not implemented. (S<Mac OS>)
1485 Implemented via Spawn. (VM/ESA)
1487 Does not automatically flush output handles on some platforms.
1488 (SunOS, Solaris, HP-UX)
1494 Emulates UNIX exit() (which considers C<exit 1> to indicate an error) by
1495 mapping the C<1> to SS$_ABORT (C<44>). This behavior may be overridden
1496 with the pragma C<use vmsish 'exit'>. As with the CRTL's exit()
1497 function, C<exit 0> is also mapped to an exit status of SS$_NORMAL
1498 (C<1>); this mapping cannot be overridden. Any other argument to exit()
1499 is used directly as Perl's exit status. (VMS)
1501 =item fcntl FILEHANDLE,FUNCTION,SCALAR
1503 Not implemented. (Win32, VMS)
1505 =item flock FILEHANDLE,OPERATION
1507 Not implemented (S<Mac OS>, VMS, S<RISC OS>, VOS).
1509 Available only on Windows NT (not on Windows 95). (Win32)
1513 Not implemented. (S<Mac OS>, AmigaOS, S<RISC OS>, VOS, VM/ESA)
1515 Emulated using multiple interpreters. See L<perlfork>. (Win32)
1517 Does not automatically flush output handles on some platforms.
1518 (SunOS, Solaris, HP-UX)
1522 Not implemented. (S<Mac OS>, S<RISC OS>)
1526 Not implemented. (S<Mac OS>, Win32, VMS, S<RISC OS>, VOS)
1530 Not implemented. (S<Mac OS>, Win32, VMS, S<RISC OS>)
1532 =item getpriority WHICH,WHO
1534 Not implemented. (S<Mac OS>, Win32, VMS, S<RISC OS>, VOS, VM/ESA)
1538 Not implemented. (S<Mac OS>, Win32)
1540 Not useful. (S<RISC OS>)
1544 Not implemented. (S<Mac OS>, Win32, VMS, S<RISC OS>)
1546 =item getnetbyname NAME
1548 Not implemented. (S<Mac OS>, Win32, Plan9)
1552 Not implemented. (S<Mac OS>, Win32)
1554 Not useful. (S<RISC OS>)
1558 Not implemented. (S<Mac OS>, Win32, VMS, S<RISC OS>)
1560 =item getnetbyaddr ADDR,ADDRTYPE
1562 Not implemented. (S<Mac OS>, Win32, Plan9)
1564 =item getprotobynumber NUMBER
1566 Not implemented. (S<Mac OS>)
1568 =item getservbyport PORT,PROTO
1570 Not implemented. (S<Mac OS>)
1574 Not implemented. (S<Mac OS>, Win32, VM/ESA)
1578 Not implemented. (S<Mac OS>, Win32, VMS, VM/ESA)
1582 Not implemented. (S<Mac OS>, Win32)
1586 Not implemented. (S<Mac OS>, Win32, Plan9)
1590 Not implemented. (S<Mac OS>, Win32, Plan9)
1594 Not implemented. (Win32, Plan9)
1598 Not implemented. (S<Mac OS>, Win32, S<RISC OS>)
1602 Not implemented. (S<Mac OS>, Win32, VMS, S<RISC OS>)
1604 =item sethostent STAYOPEN
1606 Not implemented. (S<Mac OS>, Win32, Plan9, S<RISC OS>)
1608 =item setnetent STAYOPEN
1610 Not implemented. (S<Mac OS>, Win32, Plan9, S<RISC OS>)
1612 =item setprotoent STAYOPEN
1614 Not implemented. (S<Mac OS>, Win32, Plan9, S<RISC OS>)
1616 =item setservent STAYOPEN
1618 Not implemented. (Plan9, Win32, S<RISC OS>)
1622 Not implemented. (S<Mac OS>, MPE/iX, VM/ESA, Win32)
1626 Not implemented. (S<Mac OS>, MPE/iX, S<RISC OS>, VM/ESA, VMS, Win32)
1630 Not implemented. (S<Mac OS>, Win32)
1634 Not implemented. (S<Mac OS>, Win32, Plan9)
1638 Not implemented. (S<Mac OS>, Win32, Plan9)
1642 Not implemented. (Plan9, Win32)
1644 =item getsockopt SOCKET,LEVEL,OPTNAME
1646 Not implemented. (Plan9)
1652 This operator is implemented via the File::Glob extension on most
1653 platforms. See L<File::Glob> for portability information.
1655 =item ioctl FILEHANDLE,FUNCTION,SCALAR
1657 Not implemented. (VMS)
1659 Available only for socket handles, and it does what the ioctlsocket() call
1660 in the Winsock API does. (Win32)
1662 Available only for socket handles. (S<RISC OS>)
1664 =item kill SIGNAL, LIST
1666 C<kill(0, LIST)> is implemented for the sake of taint checking;
1667 use with other signals is unimplemented. (S<Mac OS>)
1669 Not implemented, hence not useful for taint checking. (S<RISC OS>)
1671 C<kill()> doesn't have the semantics of C<raise()>, i.e. it doesn't send
1672 a signal to the identified process like it does on Unix platforms.
1673 Instead C<kill($sig, $pid)> terminates the process identified by $pid,
1674 and makes it exit immediately with exit status $sig. As in Unix, if
1675 $sig is 0 and the specified process exists, it returns true without
1676 actually terminating it. (Win32)
1678 =item link OLDFILE,NEWFILE
1680 Not implemented. (S<Mac OS>, MPE/iX, VMS, S<RISC OS>)
1682 Link count not updated because hard links are not quite that hard
1683 (They are sort of half-way between hard and soft links). (AmigaOS)
1685 Hard links are implemented on Win32 (Windows NT and Windows 2000)
1688 =item lstat FILEHANDLE
1694 Not implemented. (VMS, S<RISC OS>)
1696 Return values (especially for device and inode) may be bogus. (Win32)
1698 =item msgctl ID,CMD,ARG
1700 =item msgget KEY,FLAGS
1702 =item msgsnd ID,MSG,FLAGS
1704 =item msgrcv ID,VAR,SIZE,TYPE,FLAGS
1706 Not implemented. (S<Mac OS>, Win32, VMS, Plan9, S<RISC OS>, VOS)
1708 =item open FILEHANDLE,EXPR
1710 =item open FILEHANDLE
1712 The C<|> variants are supported only if ToolServer is installed.
1715 open to C<|-> and C<-|> are unsupported. (S<Mac OS>, Win32, S<RISC OS>)
1717 Opening a process does not automatically flush output handles on some
1718 platforms. (SunOS, Solaris, HP-UX)
1720 =item pipe READHANDLE,WRITEHANDLE
1722 Very limited functionality. (MiNT)
1728 Not implemented. (Win32, VMS, S<RISC OS>)
1730 =item select RBITS,WBITS,EBITS,TIMEOUT
1732 Only implemented on sockets. (Win32, VMS)
1734 Only reliable on sockets. (S<RISC OS>)
1736 Note that the C<select FILEHANDLE> form is generally portable.
1738 =item semctl ID,SEMNUM,CMD,ARG
1740 =item semget KEY,NSEMS,FLAGS
1742 =item semop KEY,OPSTRING
1744 Not implemented. (S<Mac OS>, Win32, VMS, S<RISC OS>, VOS)
1748 Not implemented. (MPE/iX, Win32)
1750 =item setpgrp PID,PGRP
1752 Not implemented. (S<Mac OS>, Win32, VMS, S<RISC OS>, VOS)
1754 =item setpriority WHICH,WHO,PRIORITY
1756 Not implemented. (S<Mac OS>, Win32, VMS, S<RISC OS>, VOS)
1760 Not implemented. (MPE/iX, Win32)
1762 =item setsockopt SOCKET,LEVEL,OPTNAME,OPTVAL
1764 Not implemented. (Plan9)
1766 =item shmctl ID,CMD,ARG
1768 =item shmget KEY,SIZE,FLAGS
1770 =item shmread ID,VAR,POS,SIZE
1772 =item shmwrite ID,STRING,POS,SIZE
1774 Not implemented. (S<Mac OS>, Win32, VMS, S<RISC OS>, VOS)
1776 =item sockatmark SOCKET
1778 A relatively recent addition to socket functions, may not
1779 be implemented even in UNIX platforms.
1781 =item socketpair SOCKET1,SOCKET2,DOMAIN,TYPE,PROTOCOL
1783 Not implemented. (Win32, VMS, S<RISC OS>, VOS, VM/ESA)
1785 =item stat FILEHANDLE
1791 Platforms that do not have rdev, blksize, or blocks will return these
1792 as '', so numeric comparison or manipulation of these fields may cause
1793 'not numeric' warnings.
1795 mtime and atime are the same thing, and ctime is creation time instead of
1796 inode change time. (S<Mac OS>).
1798 ctime not supported on UFS (S<Mac OS X>).
1800 ctime is creation time instead of inode change time (Win32).
1802 device and inode are not meaningful. (Win32)
1804 device and inode are not necessarily reliable. (VMS)
1806 mtime, atime and ctime all return the last modification time. Device and
1807 inode are not necessarily reliable. (S<RISC OS>)
1809 dev, rdev, blksize, and blocks are not available. inode is not
1810 meaningful and will differ between stat calls on the same file. (os2)
1812 some versions of cygwin when doing a stat("foo") and if not finding it
1813 may then attempt to stat("foo.exe") (Cygwin)
1815 =item symlink OLDFILE,NEWFILE
1817 Not implemented. (Win32, VMS, S<RISC OS>)
1821 Not implemented. (S<Mac OS>, Win32, VMS, S<RISC OS>, VOS, VM/ESA)
1823 =item sysopen FILEHANDLE,FILENAME,MODE,PERMS
1825 The traditional "0", "1", and "2" MODEs are implemented with different
1826 numeric values on some systems. The flags exported by C<Fcntl>
1827 (O_RDONLY, O_WRONLY, O_RDWR) should work everywhere though. (S<Mac
1828 OS>, OS/390, VM/ESA)
1832 In general, do not assume the UNIX/POSIX semantics that you can shift
1833 C<$?> right by eight to get the exit value, or that C<$? & 127>
1834 would give you the number of the signal that terminated the program,
1835 or that C<$? & 128> would test true if the program was terminated by a
1836 coredump. Instead, use the POSIX W*() interfaces: for example, use
1837 WIFEXITED($?) an WEXITVALUE($?) to test for a normal exit and the exit
1838 value, and WIFSIGNALED($?) and WTERMSIG($?) for a signal exit and the
1839 signal. Core dumping is not a portable concept, so there's no portable
1840 way to test for that.
1842 Only implemented if ToolServer is installed. (S<Mac OS>)
1844 As an optimization, may not call the command shell specified in
1845 C<$ENV{PERL5SHELL}>. C<system(1, @args)> spawns an external
1846 process and immediately returns its process designator, without
1847 waiting for it to terminate. Return value may be used subsequently
1848 in C<wait> or C<waitpid>. Failure to spawn() a subprocess is indicated
1849 by setting $? to "255 << 8". C<$?> is set in a way compatible with
1850 Unix (i.e. the exitstatus of the subprocess is obtained by "$? >> 8",
1851 as described in the documentation). (Win32)
1853 There is no shell to process metacharacters, and the native standard is
1854 to pass a command line terminated by "\n" "\r" or "\0" to the spawned
1855 program. Redirection such as C<< > foo >> is performed (if at all) by
1856 the run time library of the spawned program. C<system> I<list> will call
1857 the Unix emulation library's C<exec> emulation, which attempts to provide
1858 emulation of the stdin, stdout, stderr in force in the parent, providing
1859 the child program uses a compatible version of the emulation library.
1860 I<scalar> will call the native command line direct and no such emulation
1861 of a child Unix program will exists. Mileage B<will> vary. (S<RISC OS>)
1863 Far from being POSIX compliant. Because there may be no underlying
1864 /bin/sh tries to work around the problem by forking and execing the
1865 first token in its argument string. Handles basic redirection
1866 ("<" or ">") on its own behalf. (MiNT)
1868 Does not automatically flush output handles on some platforms.
1869 (SunOS, Solaris, HP-UX)
1871 The return value is POSIX-like (shifted up by 8 bits), which only allows
1872 room for a made-up value derived from the severity bits of the native
1873 32-bit condition code (unless overridden by C<use vmsish 'status'>).
1874 For more details see L<perlvms/$?>. (VMS)
1878 Only the first entry returned is nonzero. (S<Mac OS>)
1880 "cumulative" times will be bogus. On anything other than Windows NT
1881 or Windows 2000, "system" time will be bogus, and "user" time is
1882 actually the time returned by the clock() function in the C runtime
1885 Not useful. (S<RISC OS>)
1887 =item truncate FILEHANDLE,LENGTH
1889 =item truncate EXPR,LENGTH
1891 Not implemented. (Older versions of VMS)
1893 Truncation to zero-length only. (VOS)
1895 If a FILEHANDLE is supplied, it must be writable and opened in append
1896 mode (i.e., use C<<< open(FH, '>>filename') >>>
1897 or C<sysopen(FH,...,O_APPEND|O_RDWR)>. If a filename is supplied, it
1898 should not be held open elsewhere. (Win32)
1904 Returns undef where unavailable, as of version 5.005.
1906 C<umask> works but the correct permissions are set only when the file
1907 is finally closed. (AmigaOS)
1911 Only the modification time is updated. (S<BeOS>, S<Mac OS>, VMS, S<RISC OS>)
1913 May not behave as expected. Behavior depends on the C runtime
1914 library's implementation of utime(), and the filesystem being
1915 used. The FAT filesystem typically does not support an "access
1916 time" field, and it may limit timestamps to a granularity of
1917 two seconds. (Win32)
1921 =item waitpid PID,FLAGS
1923 Not implemented. (S<Mac OS>, VOS)
1925 Can only be applied to process handles returned for processes spawned
1926 using C<system(1, ...)> or pseudo processes created with C<fork()>. (Win32)
1928 Not useful. (S<RISC OS>)
1936 =item v1.48, 02 February 2001
1938 Various updates from perl5-porters over the past year, supported
1939 platforms update from Jarkko Hietaniemi.
1941 =item v1.47, 22 March 2000
1943 Various cleanups from Tom Christiansen, including migration of
1944 long platform listings from L<perl>.
1946 =item v1.46, 12 February 2000
1948 Updates for VOS and MPE/iX. (Peter Prymmer) Other small changes.
1950 =item v1.45, 20 December 1999
1952 Small changes from 5.005_63 distribution, more changes to EBCDIC info.
1954 =item v1.44, 19 July 1999
1956 A bunch of updates from Peter Prymmer for C<$^O> values,
1957 endianness, File::Spec, VMS, BS2000, OS/400.
1959 =item v1.43, 24 May 1999
1961 Added a lot of cleaning up from Tom Christiansen.
1963 =item v1.42, 22 May 1999
1965 Added notes about tests, sprintf/printf, and epoch offsets.
1967 =item v1.41, 19 May 1999
1969 Lots more little changes to formatting and content.
1971 Added a bunch of C<$^O> and related values
1972 for various platforms; fixed mail and web addresses, and added
1973 and changed miscellaneous notes. (Peter Prymmer)
1975 =item v1.40, 11 April 1999
1977 Miscellaneous changes.
1979 =item v1.39, 11 February 1999
1981 Changes from Jarkko and EMX URL fixes Michael Schwern. Additional
1982 note about newlines added.
1984 =item v1.38, 31 December 1998
1986 More changes from Jarkko.
1988 =item v1.37, 19 December 1998
1990 More minor changes. Merge two separate version 1.35 documents.
1992 =item v1.36, 9 September 1998
1994 Updated for Stratus VOS. Also known as version 1.35.
1996 =item v1.35, 13 August 1998
1998 Integrate more minor changes, plus addition of new sections under
1999 L<"ISSUES">: L<"Numbers endianness and Width">,
2000 L<"Character sets and character encoding">,
2001 L<"Internationalisation">.
2003 =item v1.33, 06 August 1998
2005 Integrate more minor changes.
2007 =item v1.32, 05 August 1998
2009 Integrate more minor changes.
2011 =item v1.30, 03 August 1998
2013 Major update for RISC OS, other minor changes.
2015 =item v1.23, 10 July 1998
2017 First public release with perl5.005.
2021 =head1 Supported Platforms
2023 As of early 2001 (the Perl releases 5.6.1 and 5.7.1), the following
2024 platforms are able to build Perl from the standard source code
2025 distribution available at http://www.cpan.org/src/index.html
2048 Tru64 UNIX (DEC OSF/1, Digital UNIX)
2054 1) in DOS mode either the DOS or OS/2 ports can be used
2055 2) Mac OS Classic (pre-X) is almost 5.6.1-ready; building from
2056 the source does work with 5.6.1, but additional MacOS specific
2057 source code is needed for a complete build. See the web
2058 site http://dev.macperl.org/ for more information.
2059 3) compilers: Borland, Cygwin, Mingw32 EGCS/GCC, VC++
2061 The following platforms worked for the previous releases (5.6.0 and 5.7.0),
2062 but we did not manage to test these in time for the 5.7.1 release.
2063 There is a very good chance that these will work fine with the 5.7.1.
2081 The following platform worked for the 5.005_03 major release but not
2082 for 5.6.0. Standardization on UTF-8 as the internal string
2083 representation in 5.6.0 and 5.6.1 introduced incompatibilities in this
2084 EBCDIC platform. While Perl 5.7.1 will build on this platform some
2085 regression tests may fail and the C<use utf8;> pragma typically
2086 introduces text handling errors.
2090 1) previously known as MVS, about to become z/OS.
2092 Strongly related to the OS/390 platform by also being EBCDIC-based
2093 mainframe platforms are the following platforms:
2098 These are also expected to work, albeit with no UTF-8 support, under 5.6.1
2099 for the same reasons as OS/390. Contact the mailing list perl-mvs@perl.org
2102 The following platforms have been known to build Perl from source in
2103 the past (5.005_03 and earlier), but we haven't been able to verify
2104 their status for the current release, either because the
2105 hardware/software platforms are rare or because we don't have an
2106 active champion on these platforms--or both. They used to work,
2107 though, so go ahead and try compiling them, and let perlbug@perl.org
2146 Support for the following platform is planned for a future Perl release:
2150 The following platforms have their own source code distributions and
2151 binaries available via http://www.cpan.org/ports/
2157 Tandem Guardian 5.004
2159 The following platforms have only binaries available via
2160 http://www.cpan.org/ports/index.html :
2164 Acorn RISCOS 5.005_02
2168 Although we do suggest that you always build your own Perl from
2169 the source code, both for maximal configurability and for security,
2170 in case you are in a hurry you can check
2171 http://www.cpan.org/ports/index.html for binary distributions.
2175 L<perlaix>, L<perlapollo>, L<perlamiga>, L<perlbeos>, L<perlbs200>,
2176 L<perlce>, L<perlcygwin>, L<perldgux>, L<perldos>, L<perlepoc>, L<perlebcdic>,
2177 L<perlhurd>, L<perlhpux>, L<perlmachten>, L<perlmacos>, L<perlmint>,
2178 L<perlmpeix>, L<perlnetware>, L<perlos2>, L<perlos390>, L<perlplan9>,
2179 L<perlqnx>, L<perlsolaris>, L<perltru64>, L<perlunicode>,
2180 L<perlvmesa>, L<perlvms>, L<perlvos>, L<perlwin32>, and L<Win32>.
2182 =head1 AUTHORS / CONTRIBUTORS
2184 Abigail <abigail@foad.org>,
2185 Charles Bailey <bailey@newman.upenn.edu>,
2186 Graham Barr <gbarr@pobox.com>,
2187 Tom Christiansen <tchrist@perl.com>,
2188 Nicholas Clark <nick@ccl4.org>,
2189 Thomas Dorner <Thomas.Dorner@start.de>,
2190 Andy Dougherty <doughera@lafayette.edu>,
2191 Dominic Dunlop <domo@computer.org>,
2192 Neale Ferguson <neale@vma.tabnsw.com.au>,
2193 David J. Fiander <davidf@mks.com>,
2194 Paul Green <Paul_Green@stratus.com>,
2195 M.J.T. Guy <mjtg@cam.ac.uk>,
2196 Jarkko Hietaniemi <jhi@iki.fi>,
2197 Luther Huffman <lutherh@stratcom.com>,
2198 Nick Ing-Simmons <nick@ing-simmons.net>,
2199 Andreas J. KE<ouml>nig <a.koenig@mind.de>,
2200 Markus Laker <mlaker@contax.co.uk>,
2201 Andrew M. Langmead <aml@world.std.com>,
2202 Larry Moore <ljmoore@freespace.net>,
2203 Paul Moore <Paul.Moore@uk.origin-it.com>,
2204 Chris Nandor <pudge@pobox.com>,
2205 Matthias Neeracher <neeracher@mac.com>,
2206 Philip Newton <pne@cpan.org>,
2207 Gary Ng <71564.1743@CompuServe.COM>,
2208 Tom Phoenix <rootbeer@teleport.com>,
2209 AndrE<eacute> Pirard <A.Pirard@ulg.ac.be>,
2210 Peter Prymmer <pvhp@forte.com>,
2211 Hugo van der Sanden <hv@crypt0.demon.co.uk>,
2212 Gurusamy Sarathy <gsar@activestate.com>,
2213 Paul J. Schinder <schinder@pobox.com>,
2214 Michael G Schwern <schwern@pobox.com>,
2215 Dan Sugalski <dan@sidhe.org>,
2216 Nathan Torkington <gnat@frii.com>.