3 perlport - Writing portable Perl
8 Perl runs on a variety of operating systems. While most of them share
9 a lot in common, they also have their own very particular and unique
12 This document is meant to help you to find out what constitutes portable
13 Perl code, so that once you have made your decision to write portably,
14 you know where the lines are drawn, and you can stay within them.
16 There is a tradeoff between taking full advantage of B<a> particular type
17 of computer, and taking advantage of a full B<range> of them. Naturally,
18 as you make your range bigger (and thus more diverse), the common
19 denominators drop, and you are left with fewer areas of common ground in
20 which you can operate to accomplish a particular task. Thus, when you
21 begin attacking a problem, it is important to consider which part of the
22 tradeoff curve you want to operate under. Specifically, whether it is
23 important to you that the task that you are coding needs the full
24 generality of being portable, or if it is sufficient to just get the job
25 done. This is the hardest choice to be made. The rest is easy, because
26 Perl provides lots of choices, whichever way you want to approach your
29 Looking at it another way, writing portable code is usually about
30 willfully limiting your available choices. Naturally, it takes discipline
33 Be aware of two important points:
37 =item Not all Perl programs have to be portable
39 There is no reason why 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 The vast majority of Perl B<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 I<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 that is commonly done
56 using a whole range of platforms, think in terms of 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, then you
65 may only need to consider 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 builtin 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 (E<lt>IMG SRC="yellow_sign.gif" ALT="Under Construction"E<gt>).
87 In most operating systems, lines in files are separated with newlines.
88 Just what is used as a newline may vary from OS to OS. Unix
89 traditionally uses C<\012>, one kind of Windows I/O uses C<\015\012>,
90 and S<Mac OS> uses C<\015>.
92 Perl uses C<\n> to represent the "logical" newline, where what
93 is logical may depend on the platform in use. In MacPerl, C<\n>
94 always means C<\015>. In DOSish perls, C<\n> usually means C<\012>, but
95 when accessing a file in "text" mode, STDIO translates it to (or from)
98 Due to the "text" mode translation, DOSish perls have limitations
99 of using C<seek> and C<tell> when a file is being accessed in "text"
100 mode. Specifically, if you stick to C<seek>-ing to locations you got
101 from C<tell> (and no others), you are usually free to use C<seek> and
102 C<tell> even in "text" mode. In general, using C<seek> or C<tell> or
103 other file operations that count bytes instead of characters, without
104 considering the length of C<\n>, may be non-portable. If you use
105 C<binmode> on a file, however, you can usually use C<seek> and C<tell>
106 with arbitrary values quite safely.
108 A common misconception in socket programming is that C<\n> eq C<\012>
109 everywhere. When using protocols such as common Internet protocols,
110 C<\012> and C<\015> are called for specifically, and the values of
111 the logical C<\n> and C<\r> (carriage return) are not reliable.
113 print SOCKET "Hi there, client!\r\n"; # WRONG
114 print SOCKET "Hi there, client!\015\012"; # RIGHT
116 [NOTE: this does not necessarily apply to communications that are
117 filtered by another program or module before sending to the socket; the
118 the most popular EBCDIC webserver, for instance, accepts C<\r\n>,
119 which translates those characters, along with all other
120 characters in text streams, from EBCDIC to ASCII.]
122 However, using C<\015\012> (or C<\cM\cJ>, or C<\x0D\x0A>) can be tedious
123 and unsightly, as well as confusing to those maintaining the code. As
124 such, the C<Socket> module supplies the Right Thing for those who want it.
126 use Socket qw(:DEFAULT :crlf);
127 print SOCKET "Hi there, client!$CRLF" # RIGHT
129 When reading I<from> a socket, remember that the default input record
130 separator (C<$/>) is C<\n>, but code like this should recognize C<$/> as
131 C<\012> or C<\015\012>:
139 use Socket qw(:DEFAULT :crlf);
140 local($/) = LF; # not needed if $/ is already \012
143 s/$CR?$LF/\n/; # not sure if socket uses LF or CRLF, OK
144 # s/\015?\012/\n/; # same thing
147 And this example is actually better than the previous one even for Unix
148 platforms, because now any C<\015>'s (C<\cM>'s) are stripped out
149 (and there was much rejoicing).
152 =head2 Numbers endianness and Width
154 Different CPUs store integers and floating point numbers in different
155 orders (called I<endianness>) and widths (32-bit and 64-bit being the
156 most common). This affects your programs if they attempt to transfer
157 numbers in binary format from a CPU architecture to another over some
158 channel: either 'live' via network connections or storing the numbers
159 to secondary storage such as a disk file.
161 Conflicting storage orders make utter mess out of the numbers: if a
162 little-endian host (Intel, Alpha) stores 0x12345678 (305419896 in
163 decimal), a big-endian host (Motorola, MIPS, Sparc, PA) reads it as
164 0x78563412 (2018915346 in decimal). To avoid this problem in network
165 (socket) connections use the C<pack()> and C<unpack()> formats C<"n">
166 and C<"N">, the "network" orders, they are guaranteed to be portable.
168 Different widths can cause truncation even between platforms of equal
169 endianness: the platform of shorter width loses the upper parts of the
170 number. There is no good solution for this problem except to avoid
171 transferring or storing raw binary numbers.
173 One can circumnavigate both these problems in two ways: either
174 transfer and store numbers always in text format, instead of raw
175 binary, or consider using modules like C<Data::Dumper> (included in
176 the standard distribution as of Perl 5.005) and C<Storable>.
180 Most platforms these days structure files in a hierarchical fashion.
181 So, it is reasonably safe to assume that any platform supports the
182 notion of a "path" to uniquely identify a file on the system. Just
183 how that path is actually written, differs.
185 While they are similar, file path specifications differ between Unix,
186 Windows, S<Mac OS>, OS/2, VMS, S<RISC OS> and probably others. Unix,
187 for example, is one of the few OSes that has the idea of a single root
190 VMS, Windows, and OS/2 can work similarly to Unix with C</> as path
191 separator, or in their own idiosyncratic ways (such as having several
192 root directories and various "unrooted" device files such NIL: and
195 S<Mac OS> uses C<:> as a path separator instead of C</>.
197 C<RISC OS> perl can emulate Unix filenames with C</> as path
198 separator, or go native and use C<.> for path separator and C<:> to
199 signal filing systems and disc names.
201 As with the newline problem above, there are modules that can help. The
202 C<File::Spec> modules provide methods to do the Right Thing on whatever
203 platform happens to be running the program.
206 chdir(File::Spec->updir()); # go up one directory
207 $file = File::Spec->catfile(
208 File::Spec->curdir(), 'temp', 'file.txt'
210 # on Unix and Win32, './temp/file.txt'
211 # on Mac OS, ':temp:file.txt'
213 File::Spec is available in the standard distribution, as of version
216 In general, production code should not have file paths hardcoded; making
217 them user supplied or from a configuration file is better, keeping in mind
218 that file path syntax varies on different machines.
220 This is especially noticeable in scripts like Makefiles and test suites,
221 which often assume C</> as a path separator for subdirectories.
223 Also of use is C<File::Basename>, from the standard distribution, which
224 splits a pathname into pieces (base filename, full path to directory,
227 Even when on a single platform (if you can call UNIX a single
228 platform), remember not to count on the existence or the contents of
229 system-specific files, like F</etc/passwd>, F</etc/sendmail.conf>, or
230 F</etc/resolv.conf>. For example the F</etc/passwd> may exist but it
231 may not contain the encrypted passwords because the system is using
232 some form of enhanced security-- or it may not contain all the
233 accounts because the system is using NIS. If code does need to rely
234 on such a file, include a description of the file and its format in
235 the code's documentation, and make it easy for the user to override
236 the default location of the file.
238 Do not have two files of the same name with different case, like
239 F<test.pl> and <Test.pl>, as many platforms have case-insensitive
240 filenames. Also, try not to have non-word characters (except for C<.>)
241 in the names, and keep them to the 8.3 convention, for maximum
244 Likewise, if using C<AutoSplit>, try to keep the split functions to
245 8.3 naming and case-insensitive conventions; or, at the very least,
246 make it so the resulting files have a unique (case-insensitively)
249 Don't assume C<E<lt>> won't be the first character of a filename. Always
250 use C<E<gt>> explicitly to open a file for reading:
252 open(FILE, "<$existing_file") or die $!;
255 =head2 System Interaction
257 Not all platforms provide for the notion of a command line, necessarily.
258 These are usually platforms that rely on a Graphical User Interface (GUI)
259 for user interaction. So a program requiring command lines might not work
260 everywhere. But this is probably for the user of the program to deal
263 Some platforms can't delete or rename files that are being held open by
264 the system. Remember to C<close> files when you are done with them.
265 Don't C<unlink> or C<rename> an open file. Don't C<tie> to or C<open> a
266 file that is already tied to or opened; C<untie> or C<close> first.
268 Don't open the same file more than once at a time for writing, as some
269 operating systems put mandatory locks on such files.
271 Don't count on a specific environment variable existing in C<%ENV>.
272 Don't count on C<%ENV> entries being case-sensitive, or even
275 Don't count on signals.
277 Don't count on filename globbing. Use C<opendir>, C<readdir>, and
280 Don't count on per-program environment variables, or per-program current
284 =head2 Interprocess Communication (IPC)
286 In general, don't directly access the system in code that is meant to be
287 portable. That means, no C<system>, C<exec>, C<fork>, C<pipe>, C<``>,
288 C<qx//>, C<open> with a C<|>, nor any of the other things that makes being
289 a Unix perl hacker worth being.
291 Commands that launch external processes are generally supported on
292 most platforms (though many of them do not support any type of forking),
293 but the problem with using them arises from what you invoke with them.
294 External tools are often named differently on different platforms, often
295 not available in the same location, often accept different arguments,
296 often behave differently, and often represent their results in a
297 platform-dependent way. Thus you should seldom depend on them to produce
300 One especially common bit of Perl code is opening a pipe to sendmail:
302 open(MAIL, '|/usr/lib/sendmail -t') or die $!;
304 This is fine for systems programming when sendmail is known to be
305 available. But it is not fine for many non-Unix systems, and even
306 some Unix systems that may not have sendmail installed. If a portable
307 solution is needed, see the C<Mail::Send> and C<Mail::Mailer> modules
308 in the C<MailTools> distribution. C<Mail::Mailer> provides several
309 mailing methods, including mail, sendmail, and direct SMTP
310 (via C<Net::SMTP>) if a mail transfer agent is not available.
312 The rule of thumb for portable code is: Do it all in portable Perl, or
313 use a module (that may internally implement it with platform-specific
314 code, but expose a common interface).
316 The UNIX System V IPC (C<msg*(), sem*(), shm*()>) is not available
317 even in all UNIX platforms.
319 =head2 External Subroutines (XS)
321 XS code, in general, can be made to work with any platform; but dependent
322 libraries, header files, etc., might not be readily available or
323 portable, or the XS code itself might be platform-specific, just as Perl
324 code might be. If the libraries and headers are portable, then it is
325 normally reasonable to make sure the XS code is portable, too.
327 There is a different kind of portability issue with writing XS
328 code: availability of a C compiler on the end-user's system. C brings
329 with it its own portability issues, and writing XS code will expose you to
330 some of those. Writing purely in perl is a comparatively easier way to
334 =head2 Standard Modules
336 In general, the standard modules work across platforms. Notable
337 exceptions are C<CPAN.pm> (which currently makes connections to external
338 programs that may not be available), platform-specific modules (like
339 C<ExtUtils::MM_VMS>), and DBM modules.
341 There is no one DBM module that is available on all platforms.
342 C<SDBM_File> and the others are generally available on all Unix and DOSish
343 ports, but not in MacPerl, where only C<NBDM_File> and C<DB_File> are
346 The good news is that at least some DBM module should be available, and
347 C<AnyDBM_File> will use whichever module it can find. Of course, then
348 the code needs to be fairly strict, dropping to the lowest common
349 denominator (e.g., not exceeding 1K for each record).
354 The system's notion of time of day and calendar date is controlled in
355 widely different ways. Don't assume the timezone is stored in C<$ENV{TZ}>,
356 and even if it is, don't assume that you can control the timezone through
359 Don't assume that the epoch starts at 00:00:00, January 1, 1970,
360 because that is OS-specific. Better to store a date in an unambiguous
361 representation. The ISO 8601 standard defines YYYY-MM-DD as the date
362 format. A text representation (like C<1 Jan 1970>) can be easily
363 converted into an OS-specific value using a module like
364 C<Date::Parse>. An array of values, such as those returned by
365 C<localtime>, can be converted to an OS-specific representation using
369 =head2 Character sets and character encoding
371 Assume very little about character sets. Do not assume anything about
372 the numerical values (C<ord()>, C<chr()>) of characters. Do not
373 assume that the alphabetic characters are encoded contiguously (in
374 numerical sense). Do no assume anything about the ordering of the
375 characters. The lowercase letters may come before or after the
376 uppercase letters, the lowercase and uppercase may be interlaced so
377 that both 'a' and 'A* come before the 'b', the accented and other
378 international characters may be interlaced so that E<auml> comes
382 =head2 Internationalisation
384 If you may assume POSIX (a rather large assumption, that: in practise
385 that means UNIX) you may read more about the POSIX locale system from
386 L<perllocale>. The locale system at least attempts to make things a
387 little bit more portable or at least more convenient and
388 native-friendly for non-English users. The system affects character
389 sets and encoding, and date and time formatting, among other things.
392 =head2 System Resources
394 If your code is destined for systems with severely constrained (or
395 missing!) virtual memory systems then you want to be I<especially> mindful
396 of avoiding wasteful constructs such as:
398 # NOTE: this is no longer "bad" in perl5.005
399 for (0..10000000) {} # bad
400 for (my $x = 0; $x <= 10000000; ++$x) {} # good
402 @lines = <VERY_LARGE_FILE>; # bad
404 while (<FILE>) {$file .= $_} # sometimes bad
405 $file = join('', <FILE>); # better
407 The last two may appear unintuitive to most people. The first of those
408 two constructs repeatedly grows a string, while the second allocates a
409 large chunk of memory in one go. On some systems, the latter is more
410 efficient that the former.
415 Most multi-user platforms provide basic levels of security that is usually
416 felt at the file-system level. Other platforms usually don't
417 (unfortunately). Thus the notion of user id, or "home" directory, or even
418 the state of being logged-in, may be unrecognizable on many platforms. If
419 you write programs that are security conscious, it is usually best to know
420 what type of system you will be operating under, and write code explicitly
421 for that platform (or class of platforms).
426 For those times when it is necessary to have platform-specific code,
427 consider keeping the platform-specific code in one place, making porting
428 to other platforms easier. Use the C<Config> module and the special
429 variable C<$^O> to differentiate platforms, as described in
435 Modules uploaded to CPAN are tested by a variety of volunteers on
436 different platforms. These CPAN testers are notified by mail of each
437 new upload, and reply to the list with PASS, FAIL, NA (not applicable to
438 this platform), or UNKNOWN (unknown), along with any relevant notations.
440 The purpose of the testing is twofold: one, to help developers fix any
441 problems in their code that crop up because of lack of testing on other
442 platforms; two, to provide users with information about whether or not
443 a given module works on a given platform.
447 =item Mailing list: cpan-testers@perl.org
449 =item Testing results: C<http://www.connect.net/gbarr/cpan-test/>
456 As of version 5.002, Perl is built with a C<$^O> variable that
457 indicates the operating system it was built on. This was implemented
458 to help speed up code that would otherwise have to C<use Config;> and
459 use the value of C<$Config{'osname'}>. Of course, to get
460 detailed information about the system, looking into C<%Config> is
461 certainly recommended.
465 Perl works on a bewildering variety of Unix and Unix-like platforms (see
466 e.g. most of the files in the F<hints/> directory in the source code kit).
467 On most of these systems, the value of C<$^O> (hence C<$Config{'osname'}>,
468 too) is determined by lowercasing and stripping punctuation from the first
469 field of the string returned by typing C<uname -a> (or a similar command)
470 at the shell prompt. Here, for example, are a few of the more popular
473 uname $^O $Config{'archname'}
474 -------------------------------------------
476 FreeBSD freebsd freebsd-i386
477 Linux linux i386-linux
478 HP-UX hpux PA-RISC1.1
480 OSF1 dec_osf alpha-dec_osf
481 SunOS solaris sun4-solaris
482 SunOS solaris i86pc-solaris
483 SunOS4 sunos sun4-sunos
485 Note that because the C<$Config{'archname'}> may depend on the hardware
486 architecture it may vary quite a lot, much more than the C<$^O>.
488 =head2 DOS and Derivatives
490 Perl has long been ported to PC style microcomputers running under
491 systems like PC-DOS, MS-DOS, OS/2, and most Windows platforms you can
492 bring yourself to mention (except for Windows CE, if you count that).
493 Users familiar with I<COMMAND.COM> and/or I<CMD.EXE> style shells should
494 be aware that each of these file specifications may have subtle
497 $filespec0 = "c:/foo/bar/file.txt";
498 $filespec1 = "c:\\foo\\bar\\file.txt";
499 $filespec2 = 'c:\foo\bar\file.txt';
500 $filespec3 = 'c:\\foo\\bar\\file.txt';
502 System calls accept either C</> or C<\> as the path separator. However,
503 many command-line utilities of DOS vintage treat C</> as the option
504 prefix, so they may get confused by filenames containing C</>. Aside
505 from calling any external programs, C</> will work just fine, and
506 probably better, as it is more consistent with popular usage, and avoids
507 the problem of remembering what to backwhack and what not to.
509 The DOS FAT filesystem can only accommodate "8.3" style filenames. Under
510 the "case insensitive, but case preserving" HPFS (OS/2) and NTFS (NT)
511 filesystems you may have to be careful about case returned with functions
512 like C<readdir> or used with functions like C<open> or C<opendir>.
514 DOS also treats several filenames as special, such as AUX, PRN, NUL, CON,
515 COM1, LPT1, LPT2 etc. Unfortunately these filenames won't even work
516 if you include an explicit directory prefix, in some cases. It is best
517 to avoid such filenames, if you want your code to be portable to DOS
520 Users of these operating systems may also wish to make use of
521 scripts such as I<pl2bat.bat> or I<pl2cmd> as appropriate to
522 put wrappers around your scripts.
524 Newline (C<\n>) is translated as C<\015\012> by STDIO when reading from
525 and writing to files. C<binmode(FILEHANDLE)> will keep C<\n> translated
526 as C<\012> for that filehandle. Since it is a noop on other systems,
527 C<binmode> should be used for cross-platform code that deals with binary
530 The C<$^O> variable and the C<$Config{'archname'}> values for various
531 DOSish perls are as follows:
533 OS $^O $Config{'archname'}
534 --------------------------------------------
538 Windows 95 MSWin32 MSWin32-x86
539 Windows NT MSWin32 MSWin32-x86
540 Windows NT MSWin32 MSWin32-alpha
541 Windows NT MSWin32 MSWin32-ppc
547 =item The djgpp environment for DOS, C<http://www.delorie.com/djgpp/>
549 =item The EMX environment for DOS, OS/2, etc. C<emx@iaehv.nl>,
550 C<http://www.juge.com/bbs/Hobb.19.html>
552 =item Build instructions for Win32, L<perlwin32>.
554 =item The ActiveState Pages, C<http://www.activestate.com/>
561 Any module requiring XS compilation is right out for most people, because
562 MacPerl is built using non-free (and non-cheap!) compilers. Some XS
563 modules that can work with MacPerl are built and distributed in binary
564 form on CPAN. See I<MacPerl: Power and Ease> and L<"CPAN Testers">
567 Directories are specified as:
569 volume:folder:file for absolute pathnames
570 volume:folder: for absolute pathnames
571 :folder:file for relative pathnames
572 :folder: for relative pathnames
573 :file for relative pathnames
574 file for relative pathnames
576 Files in a directory are stored in alphabetical order. Filenames are
577 limited to 31 characters, and may include any character except C<:>,
578 which is reserved as a path separator.
580 Instead of C<flock>, see C<FSpSetFLock> and C<FSpRstFLock> in the
581 C<Mac::Files> module.
583 In the MacPerl application, you can't run a program from the command line;
584 programs that expect C<@ARGV> to be populated can be edited with something
585 like the following, which brings up a dialog box asking for the command
589 @ARGV = split /\s+/, MacPerl::Ask('Arguments?');
592 A MacPerl script saved as a droplet will populate C<@ARGV> with the full
593 pathnames of the files dropped onto the script.
595 Mac users can use programs on a kind of command line under MPW (Macintosh
596 Programmer's Workshop, a free development environment from Apple).
597 MacPerl was first introduced as an MPW tool, and MPW can be used like a
600 perl myscript.plx some arguments
602 ToolServer is another app from Apple that provides access to MPW tools
603 from MPW and the MacPerl app, which allows MacPerl programs to use
604 C<system>, backticks, and piped C<open>.
606 "S<Mac OS>" is the proper name for the operating system, but the value
607 in C<$^O> is "MacOS". To determine architecture, version, or whether
608 the application or MPW tool version is running, check:
610 $is_app = $MacPerl::Version =~ /App/;
611 $is_tool = $MacPerl::Version =~ /MPW/;
612 ($version) = $MacPerl::Version =~ /^(\S+)/;
613 $is_ppc = $MacPerl::Architecture eq 'MacPPC';
614 $is_68k = $MacPerl::Architecture eq 'Mac68K';
616 S<Mac OS X>, to be based on NeXT's OpenStep OS, will be able to run
617 MacPerl natively (in the Blue Box, and even in the Yellow Box, once some
618 changes to the toolbox calls are made), but Unix perl will also run
625 =item The MacPerl Pages, C<http://www.ptf.com/macperl/>.
627 =item The MacPerl mailing list, C<mac-perl-request@iis.ee.ethz.ch>.
634 Perl on VMS is discussed in F<vms/perlvms.pod> in the perl distribution.
635 Note that perl on VMS can accept either VMS- or Unix-style file
636 specifications as in either of the following:
638 $ perl -ne "print if /perl_setup/i" SYS$LOGIN:LOGIN.COM
639 $ perl -ne "print if /perl_setup/i" /sys$login/login.com
641 but not a mixture of both as in:
643 $ perl -ne "print if /perl_setup/i" sys$login:/login.com
644 Can't open sys$login:/login.com: file specification syntax error
646 Interacting with Perl from the Digital Command Language (DCL) shell
647 often requires a different set of quotation marks than Unix shells do.
650 $ perl -e "print ""Hello, world.\n"""
653 There are a number of ways to wrap your perl scripts in DCL .COM files if
654 you are so inclined. For example:
656 $ write sys$output "Hello from DCL!"
658 $ then perl -x 'f$environment("PROCEDURE")
659 $ else perl -x - 'p1 'p2 'p3 'p4 'p5 'p6 'p7 'p8
660 $ deck/dollars="__END__"
663 print "Hello from Perl!\n";
668 Do take care with C<$ ASSIGN/nolog/user SYS$COMMAND: SYS$INPUT> if your
669 perl-in-DCL script expects to do things like C<$read = E<lt>STDINE<gt>;>.
671 Filenames are in the format "name.extension;version". The maximum
672 length for filenames is 39 characters, and the maximum length for
673 extensions is also 39 characters. Version is a number from 1 to
674 32767. Valid characters are C</[A-Z0-9$_-]/>.
676 VMS' RMS filesystem is case insensitive and does not preserve case.
677 C<readdir> returns lowercased filenames, but specifying a file for
678 opening remains case insensitive. Files without extensions have a
679 trailing period on them, so doing a C<readdir> with a file named F<A.;5>
680 will return F<a.> (though that file could be opened with
683 RMS had an eight level limit on directory depths from any rooted logical
684 (allowing 16 levels overall) prior to VMS 7.2. Hence
685 C<PERL_ROOT:[LIB.2.3.4.5.6.7.8]> is a valid directory specification but
686 C<PERL_ROOT:[LIB.2.3.4.5.6.7.8.9]> is not. F<Makefile.PL> authors might
687 have to take this into account, but at least they can refer to the former
688 as C</PERL_ROOT/lib/2/3/4/5/6/7/8/>.
690 The C<VMS::Filespec> module, which gets installed as part of the build
691 process on VMS, is a pure Perl module that can easily be installed on
692 non-VMS platforms and can be helpful for conversions to and from RMS
695 What C<\n> represents depends on the type of file that is open. It could
696 be C<\015>, C<\012>, C<\015\012>, or nothing. Reading from a file
697 translates newlines to C<\012>, unless C<binmode> was executed on that
698 handle, just like DOSish perls.
700 TCP/IP stacks are optional on VMS, so socket routines might not be
701 implemented. UDP sockets may not be supported.
703 The value of C<$^O> on OpenVMS is "VMS". To determine the architecture
704 that you are running on without resorting to loading all of C<%Config>
705 you can examine the content of the C<@INC> array like so:
707 if (grep(/VMS_AXP/, @INC)) {
708 print "I'm on Alpha!\n";
709 } elsif (grep(/VMS_VAX/, @INC)) {
710 print "I'm on VAX!\n";
712 print "I'm not so sure about where $^O is...\n";
721 =item vmsperl list, C<vmsperl-request@newman.upenn.edu>
723 Put words C<SUBSCRIBE VMSPERL> in message body.
725 =item vmsperl on the web, C<http://www.sidhe.org/vmsperl/index.html>
730 =head2 EBCDIC Platforms
732 Recent versions of Perl have been ported to platforms such as OS/400 on
733 AS/400 minicomputers as well as OS/390 for IBM Mainframes. Such computers
734 use EBCDIC character sets internally (usually Character Code Set ID 00819
735 for OS/400 and IBM-1047 for OS/390). Note that on the mainframe perl
736 currently works under the "Unix system services for OS/390" (formerly
737 known as OpenEdition).
739 As of R2.5 of USS for OS/390 that Unix sub-system did not support the
740 C<#!> shebang trick for script invocation. Hence, on OS/390 perl scripts
741 can executed with a header similar to the following simple script:
744 eval 'exec /usr/local/bin/perl -S $0 ${1+"$@"}'
746 #!/usr/local/bin/perl # just a comment really
748 print "Hello from perl!\n";
750 On these platforms, bear in mind that the EBCDIC character set may have
751 an effect on what happens with some perl functions (such as C<chr>,
752 C<pack>, C<print>, C<printf>, C<ord>, C<sort>, C<sprintf>, C<unpack>), as
753 well as bit-fiddling with ASCII constants using operators like C<^>, C<&>
754 and C<|>, not to mention dealing with socket interfaces to ASCII computers
757 Fortunately, most web servers for the mainframe will correctly translate
758 the C<\n> in the following statement to its ASCII equivalent (note that
759 C<\r> is the same under both Unix and OS/390):
761 print "Content-type: text/html\r\n\r\n";
763 The value of C<$^O> on OS/390 is "os390".
765 Some simple tricks for determining if you are running on an EBCDIC
766 platform could include any of the following (perhaps all):
768 if ("\t" eq "\05") { print "EBCDIC may be spoken here!\n"; }
770 if (ord('A') == 193) { print "EBCDIC may be spoken here!\n"; }
772 if (chr(169) eq 'z') { print "EBCDIC may be spoken here!\n"; }
774 Note that one thing you may not want to rely on is the EBCDIC encoding
775 of punctuation characters since these may differ from code page to code
776 page (and once your module or script is rumoured to work with EBCDIC,
777 folks will want it to work with all EBCDIC character sets).
785 The perl-mvs@perl.org list is for discussion of porting issues as well as
786 general usage issues for all EBCDIC Perls. Send a message body of
787 "subscribe perl-mvs" to majordomo@perl.org.
789 =item AS/400 Perl information at C<http://as400.rochester.ibm.com/>
796 As Acorns use ASCII with newlines (C<\n>) in text files as C<\012> like
797 Unix and Unix filename emulation is turned on by default, it is quite
798 likely that most simple scripts will work "out of the box". The native
799 filing system is modular, and individual filing systems are free to be
800 case-sensitive or insensitive, and are usually case-preserving. Some
801 native filing systems have name length limits which file and directory
802 names are silently truncated to fit - scripts should be aware that the
803 standard disc filing system currently has a name length limit of B<10>
804 characters, with up to 77 items in a directory, but other filing systems
805 may not impose such limitations.
807 Native filenames are of the form
809 Filesystem#Special_Field::DiscName.$.Directory.Directory.File
813 Special_Field is not usually present, but may contain . and $ .
814 Filesystem =~ m|[A-Za-z0-9_]|
815 DsicName =~ m|[A-Za-z0-9_/]|
816 $ represents the root directory
817 . is the path separator
818 @ is the current directory (per filesystem but machine global)
819 ^ is the parent directory
820 Directory and File =~ m|[^\0- "\.\$\%\&:\@\\^\|\177]+|
822 The default filename translation is roughly C<tr|/.|./|;>
824 Note that C<"ADFS::HardDisc.$.File" ne 'ADFS::HardDisc.$.File'> and that
825 the second stage of C<$> interpolation in regular expressions will fall
826 foul of the C<$.> if scripts are not careful.
828 Logical paths specified by system variables containing comma-separated
829 search lists are also allowed, hence C<System:Modules> is a valid
830 filename, and the filesystem will prefix C<Modules> with each section of
831 C<System$Path> until a name is made that points to an object on disc.
832 Writing to a new file C<System:Modules> would only be allowed if
833 C<System$Path> contains a single item list. The filesystem will also
834 expand system variables in filenames if enclosed in angle brackets, so
835 C<E<lt>System$DirE<gt>.Modules> would look for the file
836 S<C<$ENV{'System$Dir'} . 'Modules'>>. The obvious implication of this is
837 that B<fully qualified filenames can start with C<E<lt>E<gt>> and should
838 be protected when C<open> is used for input.
840 Because C<.> was in use as a directory separator and filenames could not
841 be assumed to be unique after 10 characters, Acorn implemented the C
842 compiler to strip the trailing C<.c> C<.h> C<.s> and C<.o> suffix from
843 filenames specified in source code and store the respective files in
844 subdirectories named after the suffix. Hence files are translated:
847 C:foo.h C:h.foo (logical path variable)
848 sys/os.h sys.h.os (C compiler groks Unix-speak)
849 10charname.c c.10charname
850 10charname.o o.10charname
851 11charname_.c c.11charname (assuming filesystem truncates at 10)
853 The Unix emulation library's translation of filenames to native assumes
854 that this sort of translation is required, and allows a user defined list
855 of known suffixes which it will transpose in this fashion. This may
856 appear transparent, but consider that with these rules C<foo/bar/baz.h>
857 and C<foo/bar/h/baz> both map to C<foo.bar.h.baz>, and that C<readdir> and
858 C<glob> cannot and do not attempt to emulate the reverse mapping. Other
859 C<.>s in filenames are translated to C</>.
861 As implied above the environment accessed through C<%ENV> is global, and
862 the convention is that program specific environment variables are of the
863 form C<Program$Name>. Each filing system maintains a current directory,
864 and the current filing system's current directory is the B<global> current
865 directory. Consequently, sociable scripts don't change the current
866 directory but rely on full pathnames, and scripts (and Makefiles) cannot
867 assume that they can spawn a child process which can change the current
868 directory without affecting its parent (and everyone else for that
871 As native operating system filehandles are global and currently are
872 allocated down from 255, with 0 being a reserved value the Unix emulation
873 library emulates Unix filehandles. Consequently, you can't rely on
874 passing C<STDIN>, C<STDOUT>, or C<STDERR> to your children.
876 The desire of users to express filenames of the form
877 C<E<lt>Foo$DirE<gt>.Bar> on the command line unquoted causes problems,
878 too: C<``> command output capture has to perform a guessing game. It
879 assumes that a string C<E<lt>[^E<lt>E<gt>]+\$[^E<lt>E<gt>]E<gt>> is a
880 reference to an environment variable, whereas anything else involving
881 C<E<lt>> or C<E<gt>> is redirection, and generally manages to be 99%
882 right. Of course, the problem remains that scripts cannot rely on any
883 Unix tools being available, or that any tools found have Unix-like command
886 Extensions and XS are, in theory, buildable by anyone using free tools.
887 In practice, many don't, as users of the Acorn platform are used to binary
888 distribution. MakeMaker does run, but no available make currently copes
889 with MakeMaker's makefiles; even if/when this is fixed, the lack of a
890 Unix-like shell can cause problems with makefile rules, especially lines
891 of the form C<cd sdbm && make all>, and anything using quoting.
893 "S<RISC OS>" is the proper name for the operating system, but the value
894 in C<$^O> is "riscos" (because we don't like shouting).
907 Perl has been ported to a variety of platforms that do not fit into any of
908 the above categories. Some, such as AmigaOS, BeOS, QNX, and Plan 9, have
909 been well-integrated into the standard Perl source code kit. You may need
910 to see the F<ports/> directory on CPAN for information, and possibly
911 binaries, for the likes of: aos, atari, lynxos, riscos, Tandem Guardian,
912 vos, I<etc.> (yes we know that some of these OSes may fall under the Unix
913 category, but we are not a standards body.)
919 =item Atari, Guido Flohr's page C<http://stud.uni-sb.de/~gufl0000/>
921 =item HP 300 MPE/iX C<http://www.cccd.edu/~markb/perlix.html>
925 A free perl5-based PERL.NLM for Novell Netware is available from
926 C<http://www.novell.com/>
931 =head1 FUNCTION IMPLEMENTATIONS
933 Listed below are functions unimplemented or implemented differently on
934 various platforms. Following each description will be, in parentheses, a
935 list of platforms that the description applies to.
937 The list may very well be incomplete, or wrong in some places. When in
938 doubt, consult the platform-specific README files in the Perl source
939 distribution, and other documentation resources for a given port.
941 Be aware, moreover, that even among Unix-ish systems there are variations.
943 For many functions, you can also query C<%Config>, exported by default
944 from C<Config.pm>. For example, to check if the platform has the C<lstat>
945 call, check C<$Config{'d_lstat'}>. See L<Config.pm> for a full
946 description of available variables.
949 =head2 Alphabetical Listing of Perl Functions
959 C<-r>, C<-w>, and C<-x> have only a very limited meaning; directories
960 and applications are executable, and there are no uid/gid
961 considerations. C<-o> is not supported. (S<Mac OS>)
963 C<-r>, C<-w>, C<-x>, and C<-o> tell whether or not file is accessible,
964 which may not reflect UIC-based file protections. (VMS)
966 C<-s> returns the size of the data fork, not the total size of data fork
967 plus resource fork. (S<Mac OS>).
969 C<-s> by name on an open file will return the space reserved on disk,
970 rather than the current extent. C<-s> on an open filehandle returns the
971 current size. (S<RISC OS>)
973 C<-R>, C<-W>, C<-X>, C<-O> are indistinguishable from C<-r>, C<-w>,
974 C<-x>, C<-o>. (S<Mac OS>, Win32, VMS, S<RISC OS>)
976 C<-b>, C<-c>, C<-k>, C<-g>, C<-p>, C<-u>, C<-A> are not implemented.
979 C<-g>, C<-k>, C<-l>, C<-p>, C<-u>, C<-A> are not particularly meaningful.
980 (Win32, VMS, S<RISC OS>)
982 C<-d> is true if passed a device spec without an explicit directory.
985 C<-T> and C<-B> are implemented, but might misclassify Mac text files
986 with foreign characters; this is the case will all platforms, but may
987 affect S<Mac OS> often. (S<Mac OS>)
989 C<-x> (or C<-X>) determine if a file ends in one of the executable
990 suffixes. C<-S> is meaningless. (Win32)
992 C<-x> (or C<-X>) determine if a file has an executable file type.
995 =item binmode FILEHANDLE
997 Meaningless. (S<Mac OS>, S<RISC OS>)
999 Reopens file and restores pointer; if function fails, underlying
1000 filehandle may be closed, or pointer may be in a different position.
1003 The value returned by C<tell> may be affected after the call, and
1004 the filehandle may be flushed. (Win32)
1008 Only limited meaning. Disabling/enabling write permission is mapped to
1009 locking/unlocking the file. (S<Mac OS>)
1011 Only good for changing "owner" read-write access, "group", and "other"
1012 bits are meaningless. (Win32)
1014 Only good for changing "owner" and "other" read-write access. (S<RISC OS>)
1018 Not implemented. (S<Mac OS>, Win32, Plan9, S<RISC OS>)
1020 Does nothing, but won't fail. (Win32)
1022 =item chroot FILENAME
1026 Not implemented. (S<Mac OS>, Win32, VMS, Plan9, S<RISC OS>)
1028 =item crypt PLAINTEXT,SALT
1030 May not be available if library or source was not provided when building
1035 Not implemented. (VMS, Plan9)
1037 =item dbmopen HASH,DBNAME,MODE
1039 Not implemented. (VMS, Plan9)
1043 Not useful. (S<Mac OS>, S<RISC OS>)
1045 Not implemented. (Win32)
1047 Invokes VMS debugger. (VMS)
1051 Not implemented. (S<Mac OS>)
1053 =item fcntl FILEHANDLE,FUNCTION,SCALAR
1055 Not implemented. (Win32, VMS)
1057 =item flock FILEHANDLE,OPERATION
1059 Not implemented (S<Mac OS>, VMS, S<RISC OS>).
1061 Available only on Windows NT (not on Windows 95). (Win32)
1065 Not implemented. (S<Mac OS>, Win32, AmigaOS, S<RISC OS>)
1069 Not implemented. (S<Mac OS>, S<RISC OS>)
1073 Not implemented. (S<Mac OS>, Win32, VMS, S<RISC OS>)
1077 Not implemented. (S<Mac OS>, Win32, VMS, S<RISC OS>)
1079 =item getpriority WHICH,WHO
1081 Not implemented. (S<Mac OS>, Win32, VMS, S<RISC OS>)
1085 Not implemented. (S<Mac OS>, Win32)
1087 Not useful. (S<RISC OS>)
1091 Not implemented. (S<Mac OS>, Win32, VMS, S<RISC OS>)
1093 =item getnetbyname NAME
1095 Not implemented. (S<Mac OS>, Win32, Plan9)
1099 Not implemented. (S<Mac OS>, Win32)
1101 Not useful. (S<RISC OS>)
1105 Not implemented. (S<Mac OS>, Win32, VMS, S<RISC OS>)
1107 =item getnetbyaddr ADDR,ADDRTYPE
1109 Not implemented. (S<Mac OS>, Win32, Plan9)
1111 =item getprotobynumber NUMBER
1113 Not implemented. (S<Mac OS>)
1115 =item getservbyport PORT,PROTO
1117 Not implemented. (S<Mac OS>)
1121 Not implemented. (S<Mac OS>, Win32)
1125 Not implemented. (S<Mac OS>, Win32, VMS)
1129 Not implemented. (S<Mac OS>, Win32)
1133 Not implemented. (S<Mac OS>, Win32, Plan9)
1137 Not implemented. (S<Mac OS>, Win32, Plan9)
1141 Not implemented. (Win32, Plan9)
1145 Not implemented. (S<Mac OS>, Win32, S<RISC OS>)
1149 Not implemented. (S<Mac OS>, Win32, VMS, S<RISC OS>)
1151 =item sethostent STAYOPEN
1153 Not implemented. (S<Mac OS>, Win32, Plan9, S<RISC OS>)
1155 =item setnetent STAYOPEN
1157 Not implemented. (S<Mac OS>, Win32, Plan9, S<RISC OS>)
1159 =item setprotoent STAYOPEN
1161 Not implemented. (S<Mac OS>, Win32, Plan9, S<RISC OS>)
1163 =item setservent STAYOPEN
1165 Not implemented. (Plan9, Win32, S<RISC OS>)
1169 Not implemented. (S<Mac OS>, Win32)
1173 Not implemented. (S<Mac OS>, Win32, VMS, S<RISC OS>)
1177 Not implemented. (S<Mac OS>, Win32)
1181 Not implemented. (S<Mac OS>, Win32, Plan9)
1185 Not implemented. (S<Mac OS>, Win32, Plan9)
1189 Not implemented. (Plan9, Win32)
1191 =item getsockopt SOCKET,LEVEL,OPTNAME
1193 Not implemented. (S<Mac OS>, Plan9)
1199 Globbing built-in, but only C<*> and C<?> metacharacters are supported.
1202 Features depend on external perlglob.exe or perlglob.bat. May be
1203 overridden with something like File::DosGlob, which is recommended.
1206 Globbing built-in, but only C<*> and C<?> metacharacters are supported.
1207 Globbing relies on operating system calls, which may return filenames
1208 in any order. As most filesystems are case-insensitive, even "sorted"
1209 filenames will not be in case-sensitive order. (S<RISC OS>)
1211 =item ioctl FILEHANDLE,FUNCTION,SCALAR
1213 Not implemented. (VMS)
1215 Available only for socket handles, and it does what the ioctlsocket() call
1216 in the Winsock API does. (Win32)
1218 Available only for socket handles. (S<RISC OS>)
1222 Not implemented, hence not useful for taint checking. (S<Mac OS>,
1225 Available only for process handles returned by the C<system(1, ...)>
1226 method of spawning a process. (Win32)
1228 =item link OLDFILE,NEWFILE
1230 Not implemented. (S<Mac OS>, Win32, VMS, S<RISC OS>)
1232 =item lstat FILEHANDLE
1238 Not implemented. (VMS, S<RISC OS>)
1240 Return values may be bogus. (Win32)
1242 =item msgctl ID,CMD,ARG
1244 =item msgget KEY,FLAGS
1246 =item msgsnd ID,MSG,FLAGS
1248 =item msgrcv ID,VAR,SIZE,TYPE,FLAGS
1250 Not implemented. (S<Mac OS>, Win32, VMS, Plan9, S<RISC OS>)
1252 =item open FILEHANDLE,EXPR
1254 =item open FILEHANDLE
1256 The C<|> variants are only supported if ToolServer is installed.
1259 open to C<|-> and C<-|> are unsupported. (S<Mac OS>, Win32, S<RISC OS>)
1261 =item pipe READHANDLE,WRITEHANDLE
1263 Not implemented. (S<Mac OS>)
1269 Not implemented. (Win32, VMS, S<RISC OS>)
1271 =item select RBITS,WBITS,EBITS,TIMEOUT
1273 Only implemented on sockets. (Win32)
1275 Only reliable on sockets. (S<RISC OS>)
1277 =item semctl ID,SEMNUM,CMD,ARG
1279 =item semget KEY,NSEMS,FLAGS
1281 =item semop KEY,OPSTRING
1283 Not implemented. (S<Mac OS>, Win32, VMS, S<RISC OS>)
1285 =item setpgrp PID,PGRP
1287 Not implemented. (S<Mac OS>, Win32, VMS, S<RISC OS>)
1289 =item setpriority WHICH,WHO,PRIORITY
1291 Not implemented. (S<Mac OS>, Win32, VMS, S<RISC OS>)
1293 =item setsockopt SOCKET,LEVEL,OPTNAME,OPTVAL
1295 Not implemented. (S<Mac OS>, Plan9)
1297 =item shmctl ID,CMD,ARG
1299 =item shmget KEY,SIZE,FLAGS
1301 =item shmread ID,VAR,POS,SIZE
1303 =item shmwrite ID,STRING,POS,SIZE
1305 Not implemented. (S<Mac OS>, Win32, VMS, S<RISC OS>)
1307 =item socketpair SOCKET1,SOCKET2,DOMAIN,TYPE,PROTOCOL
1309 Not implemented. (S<Mac OS>, Win32, VMS, S<RISC OS>)
1311 =item stat FILEHANDLE
1317 mtime and atime are the same thing, and ctime is creation time instead of
1318 inode change time. (S<Mac OS>)
1320 device and inode are not meaningful. (Win32)
1322 device and inode are not necessarily reliable. (VMS)
1324 mtime, atime and ctime all return the last modification time. Device and
1325 inode are not necessarily reliable. (S<RISC OS>)
1327 =item symlink OLDFILE,NEWFILE
1329 Not implemented. (Win32, VMS, S<RISC OS>)
1333 Not implemented. (S<Mac OS>, Win32, VMS, S<RISC OS>)
1335 =item sysopen FILEHANDLE,FILENAME,MODE,PERMS
1337 The traditional "0", "1", and "2" MODEs are implemented with different
1338 numeric values on some systems. The flags exported by C<Fcntl>
1339 (O_RDONLY, O_WRONLY, O_RDWR) should work everywhere though. (S<Mac
1344 Only implemented if ToolServer is installed. (S<Mac OS>)
1346 As an optimization, may not call the command shell specified in
1347 C<$ENV{PERL5SHELL}>. C<system(1, @args)> spawns an external
1348 process and immediately returns its process designator, without
1349 waiting for it to terminate. Return value may be used subsequently
1350 in C<wait> or C<waitpid>. (Win32)
1352 There is no shell to process metacharacters, and the native standard is
1353 to pass a command line terminated by "\n" "\r" or "\0" to the spawned
1354 program. Redirection such as C<E<gt> foo> is performed (if at all) by
1355 the run time library of the spawned program. C<system> I<list> will call
1356 the Unix emulation library's C<exec> emulation, which attempts to provide
1357 emulation of the stdin, stdout, stderr in force in the parent, providing
1358 the child program uses a compatible version of the emulation library.
1359 I<scalar> will call the native command line direct and no such emulation
1360 of a child Unix program will exists. Mileage B<will> vary. (S<RISC OS>)
1364 Only the first entry returned is nonzero. (S<Mac OS>)
1366 "cumulative" times will be bogus. On anything other than Windows NT,
1367 "system" time will be bogus, and "user" time is actually the time
1368 returned by the clock() function in the C runtime library. (Win32)
1370 Not useful. (S<RISC OS>)
1372 =item truncate FILEHANDLE,LENGTH
1374 =item truncate EXPR,LENGTH
1376 Not implemented. (VMS)
1382 Returns undef where unavailable, as of version 5.005.
1386 Only the modification time is updated. (S<Mac OS>, VMS, S<RISC OS>)
1388 May not behave as expected. Behavior depends on the C runtime
1389 library's implementation of utime(), and the filesystem being
1390 used. The FAT filesystem typically does not support an "access
1391 time" field, and it may limit timestamps to a granularity of
1392 two seconds. (Win32)
1396 =item waitpid PID,FLAGS
1398 Not implemented. (S<Mac OS>)
1400 Can only be applied to process handles returned for processes spawned
1401 using C<system(1, ...)>. (Win32)
1403 Not useful. (S<RISC OS>)
1411 =item 1.33, 06 August 1998
1413 Integrate more minor changes.
1415 =item 1.32, 05 August 1998
1417 Integrate more minor changes.
1419 =item 1.30, 03 August 1998
1421 Major update for RISC OS, other minor changes.
1423 =item 1.23, 10 July 1998
1425 First public release with perl5.005.
1429 =head1 AUTHORS / CONTRIBUTORS
1431 Abigail E<lt>abigail@fnx.comE<gt>,
1432 Charles Bailey E<lt>bailey@genetics.upenn.eduE<gt>,
1433 Graham Barr E<lt>gbarr@pobox.comE<gt>,
1434 Tom Christiansen E<lt>tchrist@perl.comE<gt>,
1435 Nicholas Clark E<lt>Nicholas.Clark@liverpool.ac.ukE<gt>,
1436 Andy Dougherty E<lt>doughera@lafcol.lafayette.eduE<gt>,
1437 Dominic Dunlop E<lt>domo@vo.luE<gt>,
1438 M.J.T. Guy E<lt>mjtg@cus.cam.ac.ukE<gt>,
1439 Luther Huffman E<lt>lutherh@stratcom.comE<gt>,
1440 Nick Ing-Simmons E<lt>nick@ni-s.u-net.comE<gt>,
1441 Andreas J. KE<ouml>nig E<lt>koenig@kulturbox.deE<gt>,
1442 Andrew M. Langmead E<lt>aml@world.std.comE<gt>,
1443 Paul Moore E<lt>Paul.Moore@uk.origin-it.comE<gt>,
1444 Chris Nandor E<lt>pudge@pobox.comE<gt>,
1445 Matthias Neeracher E<lt>neeri@iis.ee.ethz.chE<gt>,
1446 Gary Ng E<lt>71564.1743@CompuServe.COME<gt>,
1447 Tom Phoenix E<lt>rootbeer@teleport.comE<gt>,
1448 Peter Prymmer E<lt>pvhp@forte.comE<gt>,
1449 Hugo van der Sanden E<lt>hv@crypt0.demon.co.ukE<gt>,
1450 Gurusamy Sarathy E<lt>gsar@umich.eduE<gt>,
1451 Paul J. Schinder E<lt>schinder@pobox.comE<gt>,
1452 Dan Sugalski E<lt>sugalskd@ous.eduE<gt>,
1453 Nathan Torkington E<lt>gnat@frii.comE<gt>.
1455 This document is maintained by Chris Nandor.
1459 Version 1.34, last modified 07 August 1998.