Seth Woolley's Man Viewer

perlport(1) - perlport - Writing portable Perl - man 1 perlport

([section] manual, -k keyword, -K [section] search, -f whatis)
man plain no title

PERLPORT(1)            Perl Programmers Reference Guide            PERLPORT(1)

       perlport - Writing portable Perl

       Perl runs on numerous operating systems.  While most of them share much
       in(1,8) common, they also have their own unique features.

       This document is meant to help you to find out what constitutes porta-
       ble Perl code.  That way once you make a decision to write(1,2) portably,
       you know where the lines are drawn, and you can stay within them.

       There is a tradeoff between taking full advantage of one particular
       type of computer and taking advantage of a full range of them.  Natu-
       rally, as you broaden your range and become more diverse, the common
       factors drop, and you are left with an increasingly smaller area of
       common ground in(1,8) which you can operate to accomplish a particular task.
       Thus, when you begin attacking a problem, it is important to consider
       under which part of the tradeoff curve you want to operate.  Specifi-
       cally, you must decide whether it is important that the task that you
       are coding have the full generality of being portable, or whether to
       just get the job done right now.  This is the hardest choice to be
       made.  The rest is easy, because Perl provides many choices, whichever
       way you want to approach your problem.

       Looking at it another way, writing portable code is usually about will-
       fully limiting your available choices.  Naturally, it takes discipline
       and sacrifice to do that.  The product of portability and convenience
       may be a constant.  You have been warned.

       Be aware of two important points:

       Not all Perl programs have to be portable
           There is no reason you should not use Perl as a language to glue
           Unix tools together, or to prototype a Macintosh application, or to
           manage the Windows registry.  If it makes no sense to aim for
           portability for one reason or another in(1,8) a given program, then
           don't bother.

       Nearly all of Perl already is portable
           Don't be fooled into thinking that it is hard to create portable
           Perl code.  It isn't.  Perl tries its level-best to bridge the gaps
           between what's available on different platforms, and all the means
           available to use those features.  Thus almost all Perl code runs on
           any machine without modification.  But there are some significant
           issues in(1,8) writing portable code, and this document is entirely
           about those issues.

       Here's the general rule: When you approach a task commonly done using a
       whole range of platforms, think about writing portable code.  That way,
       you don't sacrifice much by way of the implementation choices you can
       avail yourself of, and at the same time(1,2,n) you can give your users(1,5) lots of
       platform choices.  On the other hand, when you have to take advantage
       of some unique feature of a particular platform, as is often the case
       with systems programming (whether for Unix, Windows, Mac OS, VMS,
       etc.), consider writing platform-specific code.

       When the code will run on only two or three operating systems, you may
       need to consider only the differences of those particular systems.  The
       important thing is to decide where the code will run and to be deliber-
       ate in(1,8) your decision.

       The material below is separated into three main sections: main issues
       of portability ("ISSUES", platform-specific issues ("PLATFORMS", and
       built-in perl functions that behave differently on various ports

       This information should not be considered complete; it includes possi-
       bly transient information about idiosyncrasies of some of the ports,
       almost all of which are in(1,8) a state of constant evolution.  Thus, this
       material should be considered a perpetual work in(1,8) progress ("<IMG
       SRC="yellow_sign.gif" ALT="Under Construction">").


       In most operating systems, lines in(1,8) files are terminated by newlines.
       Just what is used as a newline may vary from OS to OS.  Unix tradition-
       ally uses "\012", one type of DOSish I/O uses "\015\012", and Mac OS
       uses "\015".

       Perl uses "\n" to represent the "logical" newline, where what is logi-
       cal may depend on the platform in(1,8) use.  In MacPerl, "\n" always means
       "\015".  In DOSish perls, "\n" usually means "\012", but when accessing
       a file(1,n) in(1,8) "text" mode, STDIO translates it to (or from) "\015\012",
       depending on whether you're reading or writing.  Unix does the same
       thing on ttys in(1,8) canonical mode.  "\015\012" is commonly referred to as

       A common cause of unportable programs is the misuse of chop() to trim

           # XXX UNPORTABLE!
           while(<FILE>) {
               @array = split(1,n)(/:/);

       You can get away with this on Unix and Mac OS (they have a single char-
       acter end-of-line), but the same program will break under DOSish perls
       because you're only chop()ing half the end-of-line.  Instead, chomp()
       should be used to trim newlines.  The Dunce::Files module can help
       audit your code for misuses of chop().

       When dealing with binary files (or text files in(1,8) binary mode) be sure
       to explicitly set(7,n,1 builtins) $/ to the appropriate value for your file(1,n) format
       before using chomp().

       Because of the "text" mode translation, DOSish perls have limitations
       in(1,8) using "seek" and "tell" on a file(1,n) accessed in(1,8) "text" mode.  Stick to
       "seek"-ing to locations you got from "tell" (and no others), and you
       are usually free to use "seek" and "tell" even in(1,8) "text" mode.  Using
       "seek" or "tell" or other file(1,n) operations may be non-portable.  If you
       use "binmode" on a file(1,n), however, you can usually "seek" and "tell"
       with arbitrary values in(1,8) safety.

       A common misconception in(1,8) socket(2,7,n) programming is that "\n" eq "\012"
       everywhere.  When using protocols such as common Internet protocols,
       "\012" and "\015" are called for specifically, and the values of the
       logical "\n" and "\r" (carriage return) are not reliable.

           print SOCKET "Hi there, client!\r\n";      # WRONG
           print SOCKET "Hi there, client!\015\012";  # RIGHT

       However, using "\015\012" (or "\cM\cJ", or "\x0D\x0A") can be tedious
       and unsightly, as well as confusing to those maintaining the code.  As
       such, the Socket module supplies the Right Thing for those who want it.

           use Socket qw(:DEFAULT :crlf);
           print SOCKET "Hi there, client!$CRLF"      # RIGHT

       When reading from a socket(2,7,n), remember that the default input record sep-
       arator $/ is "\n", but robust socket(2,7,n) code will recognize as either
       "\012" or "\015\012" as end of line:

           while (<SOCKET>) {
               # ...

       Because both CRLF and LF end in(1,8) LF, the input record separator can be
       set(7,n,1 builtins) to LF and any CR stripped later.  Better to write:

           use Socket qw(:DEFAULT :crlf);
           local($/) = LF;      # not needed if(3,n) $/ is already \012

           while (<SOCKET>) {
               s/$CR?$LF/\n/;   # not sure if(3,n) socket(2,7,n) uses LF or CRLF, OK
           #   s/\015?\012/\n/; # same thing

       This example is preferred over the previous one--even for Unix plat-
       forms--because now any "\015"'s ("\cM"'s) are stripped out (and there
       was much rejoicing).

       Similarly, functions that return text data--such as a function that
       fetches a web page--should sometimes translate newlines before return-
       ing the data, if(3,n) they've not yet been translated to the local newline
       representation.  A single line of code will often suffice:

           $data =~ s/\015?\012/\n/g;
           return $data;

       Some of this may be confusing.  Here's a handy reference to the ASCII
       CR and LF characters.  You can print it out and stick it in(1,8) your wal-

           LF  eq  \012  eq  \x0A  eq  \cJ  eq  chr(10)  eq  ASCII 10
           CR  eq  \015  eq  \x0D  eq  \cM  eq  chr(13)  eq  ASCII 13

                    | Unix | DOS  | Mac  |
               \n   |  LF  |  LF  |  CR  |
               \r   |  CR  |  CR  |  LF  |
               \n * |  LF  | CRLF |  CR  |
               \r * |  CR  |  CR  |  LF  |
               * text-mode STDIO

       The Unix column assumes that you are not accessing a serial line (like
       a tty(1,4)) in(1,8) canonical mode.  If you are, then CR on input becomes "\n",
       and "\n" on output becomes CRLF.

       These are just the most common definitions of "\n" and "\r" in(1,8) Perl.
       There may well be others.  For example, on an EBCDIC implementation
       such as z/OS (OS/390) or OS/400 (using the ILE, the PASE is
       ASCII-based) the above material is similar to "Unix" but the code num-
       bers change:

           LF  eq  \025  eq  \x15  eq  \cU  eq  chr(21)  eq  CP-1047 21
           LF  eq  \045  eq  \x25  eq           chr(37)  eq  CP-0037 37
           CR  eq  \015  eq  \x0D  eq  \cM  eq  chr(13)  eq  CP-1047 13
           CR  eq  \015  eq  \x0D  eq  \cM  eq  chr(13)  eq  CP-0037 13

                    | z/OS | OS/400 |
               \n   |  LF  |  LF    |
               \r   |  CR  |  CR    |
               \n * |  LF  |  LF    |
               \r * |  CR  |  CR    |
               * text-mode STDIO

       Numbers endianness and Width

       Different CPUs store integers and floating point numbers in(1,8) different
       orders (called endianness) and widths (32-bit and 64-bit being the most
       common today).  This affects your programs when they attempt to trans-
       fer numbers in(1,8) binary format from one CPU architecture to another, usu-
       ally either "live" via network connection, or by storing the numbers to
       secondary storage such as a disk file(1,n) or tape.

       Conflicting storage orders make utter mess out of the numbers.  If a
       little-endian host(1,5) (Intel, VAX) stores 0x12345678 (305419896 in(1,8) deci-
       mal), a big-endian host(1,5) (Motorola, Sparc, PA) reads it as 0x78563412
       (2018915346 in(1,8) decimal).  Alpha and MIPS can be either: Digital/Compaq
       used/uses them in(1,8) little-endian mode; SGI/Cray uses them in(1,8) big-endian
       mode.  To avoid this problem in(1,8) network (socket(2,7,n)) connections use the
       "pack(3,n,n pack-old)" and "unpack" formats "n" and "N", the "network" orders.  These
       are guaranteed to be portable.

       You can explore the endianness of your platform by unpacking a data
       structure packed in(1,8) native format such as:

           print unpack("h*", pack(3,n,n pack-old)("s2", 1, 2)), "\n";
           # '10002000' on e.g. Intel x86 or Alpha 21064 in(1,8) little-endian mode
           # '00100020' on e.g. Motorola 68040

       If you need to distinguish between endian architectures you could use
       either of the variables set(7,n,1 builtins) like so:

           $is_big_endian   = unpack("h*", pack(3,n,n pack-old)("s", 1)) =~ /01/;
           $is_little_endian = unpack("h*", pack(3,n,n pack-old)("s", 1)) =~ /^1/;

       Differing widths can cause truncation even between platforms of equal
       endianness.  The platform of shorter width loses the upper parts of the
       number.  There is no good solution for this problem except to avoid
       transferring or storing raw(3x,7,8,3x cbreak) binary numbers.

       One can circumnavigate both these problems in(1,8) two ways.  Either trans-
       fer and store numbers always in(1,8) text format, instead of raw(3x,7,8,3x cbreak) binary, or
       else consider using modules like Data::Dumper (included in(1,8) the standard
       distribution as of Perl 5.005) and Storable (included as of perl 5.8).
       Keeping all data as text significantly simplifies matters.

       The v-strings are portable only up to v2147483647 (0x7FFFFFFF), that's
       how far EBCDIC, or more precisely UTF-EBCDIC will go.

       Files and Filesystems

       Most platforms these days structure files in(1,8) a hierarchical fashion.
       So, it is reasonably safe to assume that all platforms support the
       notion of a "path" to uniquely identify a file(1,n) on the system.  How that
       path is really written, though, differs considerably.

       Although similar, file(1,n) path specifications differ between Unix, Win-
       dows, Mac OS, OS/2, VMS, VOS, RISC OS, and probably others.  Unix, for
       example, is one of the few OSes that has the elegant idea of a single
       root directory.

       DOS, OS/2, VMS, VOS, and Windows can work similarly to Unix with "/" as
       path separator, or in(1,8) their own idiosyncratic ways (such as having sev-
       eral root directories and various "unrooted" device files such NIL: and

       Mac OS uses ":" as a path separator instead of "/".

       The filesystem may support neither hard links ("link(1,2)") nor symbolic
       links ("symlink", "readlink(1,2)", "lstat").

       The filesystem may support neither access(2,5) timestamp nor change time-
       stamp (meaning that about the only portable timestamp is the modifica-
       tion timestamp), or one second granularity of any timestamps (e.g. the
       FAT filesystem limits the time(1,2,n) granularity to two seconds).

       The "inode change timestamp" (the "-C" filetest) may really be the
       "creation timestamp" (which it is not in(1,8) UNIX).

       VOS perl can emulate Unix filenames with "/" as path separator.  The
       native pathname characters greater-than, less-than, number-sign, and
       percent-sign are always accepted.

       RISC OS perl can emulate Unix filenames with "/" as path separator, or
       go native and use "." for path separator and ":" to signal(2,7) filesystems
       and disk names.

       Don't assume UNIX filesystem access(2,5) semantics: that read(2,n,1 builtins), write(1,2), and
       execute are all the permissions there are, and even if(3,n) they exist, that
       their semantics (for example what do r, w, and x mean on a directory)
       are the UNIX ones.  The various UNIX/POSIX compatibility layers usually
       try to make interfaces like chmod(1,2)() work, but sometimes there simply is
       no good mapping.

       If all this is intimidating, have no (well, maybe only a little) fear.
       There are modules that can help.  The File::Spec modules provide meth-
       ods to do the Right Thing on whatever platform happens to be running
       the program.

           use File::Spec::Functions;
           chdir(updir());        # go up one directory
           $file(1,n) = catfile(curdir(), 'temp', 'file.txt');
           # on Unix and Win32, './temp/file.txt'
           # on Mac OS, ':temp:file.txt'
           # on VMS, '[.temp]file.txt'

       File::Spec is available in(1,8) the standard distribution as of version(1,3,5)
       5.004_05.  File::Spec::Functions is only in(1,8) File::Spec 0.7 and later,
       and some versions of perl come with version(1,3,5) 0.6.  If File::Spec is not
       updated to 0.7 or later, you must use the object-oriented interface
       from File::Spec (or upgrade File::Spec).

       In general, production code should not have file(1,n) paths hardcoded.  Mak-
       ing them user-supplied or read(2,n,1 builtins) from a configuration file(1,n) is better,
       keeping in(1,8) mind that file(1,n) path syntax varies on different machines.

       This is especially noticeable in(1,8) scripts like Makefiles and test
       suites, which often assume "/" as a path separator for subdirectories.

       Also of use is File::Basename from the standard distribution, which
       splits a pathname into pieces (base filename, full path to directory,
       and file(1,n) suffix).

       Even when on a single platform (if(3,n) you can call Unix a single plat-
       form), remember not to count on the existence or the contents of par-
       ticular system-specific files or directories, like /etc/passwd(1,5),
       /etc/sendmail.conf, /etc/resolv.conf, or even /tmp/.  For example,
       /etc/passwd(1,5) may exist but not contain the encrypted passwords, because
       the system is using some form of enhanced security.  Or it may not con-
       tain all the accounts, because the system is using NIS.  If code does
       need to rely on such a file(1,n), include a description of the file(1,n) and its
       format in(1,8) the code's documentation, then make it easy for the user to
       override the default location of the file.

       Don't assume a text file(1,n) will end with a newline.  They should, but
       people forget.

       Do not have two files or directories of the same name with different
       case, like and, as many platforms have case-insensitive
       (or at least case-forgiving) filenames.  Also, try not to have non-word
       characters (except for ".") in(1,8) the names, and keep them to the 8.3 con-
       vention, for maximum portability, onerous a burden though this may

       Likewise, when using the AutoSplit module, try to keep your functions
       to 8.3 naming and case-insensitive conventions; or, at the least, make
       it so the resulting files have a unique (case-insensitively) first 8

       Whitespace in(1,8) filenames is tolerated on most systems, but not all, and
       even on systems where it might be tolerated, some utilities might
       become confused by such whitespace.

       Many systems (DOS, VMS) cannot have more than one "." in(1,8) their file-

       Don't assume ">" won't be the first character of a filename.  Always
       use "<" explicitly to open(2,3,n) a file(1,n) for reading, or even better, use the
       three-arg version(1,3,5) of open(2,3,n), unless you want the user to be able to spec-
       ify a pipe(2,8) open.

           open(2,3,n)(FILE, '<', $existing_file) or die $!;

       If filenames might use strange characters, it is safest to open(2,3,n) it with
       "sysopen" instead of "open(2,3,n)".  "open(2,3,n)" is magic(4,5) and can translate charac-
       ters like ">", "<", and "|", which may be the wrong thing to do.
       (Sometimes, though, it's the right thing.)  Three-arg open(2,3,n) can also
       help protect against this translation in(1,8) cases where it is undesirable.

       Don't use ":" as a part of a filename since many systems use that for
       their own semantics (Mac OS Classic for separating pathname components,
       many networking schemes and utilities for separating the nodename and
       the pathname, and so on).  For the same reasons, avoid "@", ";" and

       Don't assume that in(1,8) pathnames you can collapse two leading slashes
       "//" into one: some networking and clustering filesystems have special
       semantics for that.  Let the operating system to sort(1,3) it out.

       The portable filename characters as defined by ANSI C are

        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
        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
        0 1 2 3 4 5 6 7 8 9
        . _ -

       and the "-" shouldn't be the first character.  If you want to be hyper-
       correct, stay case-insensitive and within the 8.3 naming convention
       (all the files and directories have to be unique within one directory
       if(3,n) their names are lowercased and truncated to eight characters before
       the ".", if(3,n) any, and to three characters after the ".", if(3,n) any).  (And
       do not use "."s in(1,8) directory names.)

       System Interaction

       Not all platforms provide a command line.  These are usually platforms
       that rely primarily on a Graphical User Interface (GUI) for user inter-
       action.  A program requiring a command line interface might not work
       everywhere.  This is probably for the user of the program to deal with,
       so don't stay up late worrying about it.

       Some platforms can't delete or rename(1,2,n) files held open(2,3,n) by the system,
       this limitation may also apply to changing filesystem metainformation
       like file(1,n) permissions or owners.  Remember to "close(2,7,n)" files when you
       are done with them.  Don't "unlink(1,2)" or "rename(1,2,n)" an open(2,3,n) file.  Don't
       "tie" or "open(2,3,n)" a file(1,n) already tied or opened; "untie" or "close(2,7,n)" it

       Don't open(2,3,n) the same file(1,n) more than once at a time(1,2,n) for writing, as some
       operating systems put mandatory locks on such files.

       Don't assume that write(1,2)/modify permission on a directory gives the
       right to add or delete files/directories in(1,8) that directory.  That is
       filesystem specific: in(1,8) some filesystems you need write(1,2)/modify permis-
       sion also (or even just) in(1,8) the file(1,n)/directory itself.  In some
       filesystems (AFS, DFS) the permission to add/delete directory entries
       is a completely separate permission.

       Don't assume that a single "unlink(1,2)" completely gets(3,n) rid of the file:
       some filesystems (most notably the ones in(1,8) VMS) have versioned filesys-
       tems, and unlink(1,2)() removes only the most recent one (it doesn't remove
       all the versions because by default the native tools on those platforms
       remove just the most recent version(1,3,5), too).  The portable idiom to
       remove all the versions of a file(1,n) is

           1 while unlink(1,2) "file(1,n)";

       This will terminate if(3,n) the file(1,n) is undeleteable for some reason (pro-
       tected, not there, and so on).

       Don't count on a specific environment variable existing in(1,8) %ENV.  Don't
       count on %ENV entries being case-sensitive, or even case-preserving.
       Don't try to clear(1,3x,3x clrtobot) %ENV by saying "%ENV = ();", or, if(3,n) you really have
       to, make it conditional on "$^O ne 'VMS'" since in(1,8) VMS the %ENV table
       is much more than a per-process key-value string(3,n) table.

       Don't count on signals or %SIG for anything.

       Don't count on filename globbing.  Use "opendir", "readdir(2,3)", and
       "closedir" instead.

       Don't count on per-program environment variables, or per-program cur-
       rent directories.

       Don't count on specific values of $!, neither numeric nor especially
       the strings values-- users(1,5) may switch(1,n) their locales causing error(8,n) mes-
       sages to be translated into their languages.  If you can trust a POSIX-
       ish environment, you can portably use the symbols defined by the Errno
       module, like ENOENT.  And don't trust on the values of $!  at all
       except immediately after a failed system call.

       Command names versus file(1,n) pathnames

       Don't assume that the name used to invoke a command or program with
       "system" or "exec(3,n,1 builtins)" can also be used to test for the existence of the
       file(1,n) that holds the executable code for that command or program.
       First, many systems have "internal" commands that are built-in to the
       shell or OS and while these commands can be invoked, there is no corre-
       sponding file.  Second, some operating systems (e.g., Cygwin, DJGPP,
       OS/2, and VOS) have required suffixes for executable files; these suf-
       fixes are generally permitted on the command name but are not required.
       Thus, a command like "perl" might exist in(1,8) a file(1,n) named(5,8) "perl",
       "perl.exe", or "", depending on the operating system.  The vari-
       able "_exe" in(1,8) the Config module holds the executable suffix, if(3,n) any.
       Third, the VMS port carefully sets up $^X and $Config{perlpath} so that
       no further processing is required.  This is just as well, because the
       matching regular expression used below would then have to deal with a
       possible trailing version(1,3,5) number in(1,8) the VMS file(1,n) name.

       To convert $^X to a file(1,n) pathname, taking account of the requirements
       of the various operating system possibilities, say:
         use Config;
         $thisperl = $^X;
         if(3,n) ($^O ne 'VMS')
            {$thisperl .= $Config{_exe} unless $thisperl =~ m/$Con-

       To convert $Config{perlpath} to a file(1,n) pathname, say:
         use Config;
         $thisperl = $Config{perlpath};
         if(3,n) ($^O ne 'VMS')
            {$thisperl .= $Config{_exe} unless $thisperl =~ m/$Con-


       Don't assume that you can reach the public Internet.

       Don't assume that there is only one way to get through firewalls to the
       public Internet.

       Don't assume that you can reach outside world through any other port
       than 80, or some web proxy.  ftp is blocked by many firewalls.

       Don't assume that you can send(2,n) email by connecting to the local SMTP

       Don't assume that you can reach yourself or any node by the name
       'localhost'.  The same goes for ''.  You will have to try

       Don't assume that the host(1,5) has only one network card, or that it can't
       bind(2,n,1 builtins) to many virtual(5,8) IP addresses.

       Don't assume a particular network device name.

       Don't assume a particular set(7,n,1 builtins) of ioctl()s will work.

       Don't assume that you can ping hosts and get replies.

       Don't assume that any particular port (service) will respond.

       Don't assume that Sys::Hostname() (or any other API or command) returns
       either a fully qualified hostname or a non-qualified hostname: it all
       depends on how the system had been configured.  Also remember things
       like DHCP and NAT-- the hostname you get back might not be very useful.

       All the above "don't":s may look(1,8,3 Search::Dict) daunting, and they are -- but the key
       is to degrade gracefully if(3,n) one cannot reach the particular network
       service one wants.  Croaking or hanging do not look(1,8,3 Search::Dict) very professional.

       Interprocess Communication (IPC)

       In general, don't directly access(2,5) the system in(1,8) code meant to be porta-
       ble.  That means, no "system", "exec(3,n,1 builtins)", "fork", "pipe(2,8)", ``, "qx//",
       "open(2,3,n)" with a "|", nor any of the other things that makes being a perl
       hacker worth being.

       Commands that launch external processes are generally supported on most
       platforms (though many of them do not support any type of forking).
       The problem with using them arises from what you invoke them on.
       External tools are often named(5,8) differently on different platforms, may
       not be available in(1,8) the same location, might accept(2,8) different argu-
       ments, can behave differently, and often present their results in(1,8) a
       platform-dependent way.  Thus, you should seldom depend on them to pro-
       duce consistent results. (Then again, if(3,n) you're calling netstat -a, you
       probably don't expect it to run on both Unix and CP/M.)

       One especially common bit of Perl code is opening a pipe(2,8) to sendmail(1,8):

           open(2,3,n)(MAIL, '|/usr/lib/sendmail(1,8) -t')
               or die "cannot fork sendmail: $!";

       This is fine for systems programming when sendmail(1,8) is known to be
       available.  But it is not fine for many non-Unix systems, and even some
       Unix systems that may not have sendmail(1,8) installed.  If a portable solu-
       tion is needed, see the various distributions on CPAN that deal with
       it.  Mail::Mailer and Mail::Send in(1,8) the MailTools distribution are com-
       monly used, and provide several mailing methods, including mail(1,8), send-
       mail(1,8), and direct SMTP (via Net::SMTP) if(3,n) a mail(1,8) transfer agent is not
       available.  Mail::Sendmail is a standalone module that provides simple,
       platform-independent mailing.

       The Unix System V IPC ("msg*(), sem*(), shm*()") is not available even
       on all Unix platforms.

       Do not use either the bare result of "pack(3,n,n pack-old)("N", 10, 20, 30, 40)" or
       bare v-strings (such as "v10.20.30.40") to represent IPv4 addresses:
       both forms just pack(3,n,n pack-old) the four bytes into network order.  That this
       would be equal to the C language "in_addr" struct (which is what the
       socket(2,7,n) code internally uses) is not guaranteed.  To be portable use the
       routines of the Socket extension, such as "inet_aton()", "inet_ntoa()",
       and "sockaddr_in()".

       The rule of thumb for portable code is: Do it all in(1,8) portable Perl, or
       use a module (that may internally implement it with platform-specific
       code, but expose a common interface).

       External Subroutines (XS)

       XS code can usually be made to work with any platform, but dependent
       libraries, header files, etc., might not be readily available or porta-
       ble, or the XS code itself might be platform-specific, just as Perl
       code might be.  If the libraries and headers are portable, then it is
       normally reasonable to make sure the XS code is portable, too.

       A different type of portability issue arises when writing XS code:
       availability of a C compiler on the end-user's system.  C brings with
       it its own portability issues, and writing XS code will expose you to
       some of those.  Writing purely in(1,8) Perl is an easier way to achieve

       Standard Modules

       In general, the standard modules work across platforms.  Notable excep-
       tions are the CPAN module (which currently makes connections to exter-
       nal programs that may not be available), platform-specific modules
       (like ExtUtils::MM_VMS), and DBM modules.

       There is no one DBM module available on all platforms.  SDBM_File and
       the others are generally available on all Unix and DOSish ports, but
       not in(1,8) MacPerl, where only NBDM_File and DB_File are available.

       The good news is that at least some DBM module should be available, and
       AnyDBM_File will use whichever module it can find.  Of course, then the
       code needs to be fairly strict, dropping to the greatest common factor(1,6)
       (e.g., not exceeding 1K for each record), so that it will work with any
       DBM module.  See AnyDBM_File for more details.

       Time and Date

       The system's notion of time(1,2,n) of day and calendar date is controlled in(1,8)
       widely different ways.  Don't assume the timezone is stored in(1,8)
       $ENV{TZ}, and even if(3,n) it is, don't assume that you can control the
       timezone through that variable.  Don't assume anything about the three-
       letter timezone abbreviations (for example that MST would be the Moun-
       tain Standard Time, it's been known to stand for Moscow Standard Time).
       If you need to use timezones, express them in(1,8) some unambiguous format
       like the exact number of minutes offset from UTC, or the POSIX timezone

       Don't assume that the epoch starts at 00:00:00, January 1, 1970,
       because that is OS- and implementation-specific.  It is better to store
       a date in(1,8) an unambiguous representation.  The ISO 8601 standard defines
       YYYY-MM-DD as the date format, or YYYY-MM-DDTHH-MM-SS (that's a literal
       "T" separating the date from the time(1,2,n)).  Please do use the ISO 8601
       instead of making us to guess what date 02/03/04 might be.  ISO 8601
       even sorts nicely as-is.  A text representation (like "1987-12-18") can
       be easily converted into an OS-specific value using a module like
       Date::Parse.  An array of values, such as those returned by "local-
       time(1,2,n)", can be converted to an OS-specific representation using

       When calculating specific times, such as for tests in(1,8) time(1,2,n) or date mod-
       ules, it may be appropriate to calculate an offset for the epoch.

           require Time::Local;
           $offset = Time::Local::timegm(0, 0, 0, 1, 0, 70);

       The value for $offset in(1,8) Unix will be 0, but in(1,8) Mac OS will be some
       large number.  $offset can then be added to a Unix time(1,2,n) value to get
       what should be the proper value on any system.

       On Windows (at least), you shouldn't pass a negative value to "gmtime"
       or "localtime".

       Character sets and character encoding(3,n)

       Assume very little about character sets.

       Assume nothing about numerical values ("ord", "chr") of characters.  Do
       not use explicit code point ranges (like \xHH-\xHH); use for example
       symbolic character classes like "[:print:]".

       Do not assume that the alphabetic characters are encoded contiguously
       (in(1,8) the numeric sense).  There may be gaps.

       Do not assume anything about the ordering of the characters.  The low-
       ercase letters may come before or after the uppercase letters; the low-
       ercase and uppercase may be interlaced so that both `a' and `A' come
       before `b'; the accented and other international characters may be
       interlaced so that ae comes before `b'.


       If you may assume POSIX (a rather large assumption), you may read(2,n,1 builtins) more
       about the POSIX locale(3,5,7) system from perllocale.  The locale(3,5,7) system at
       least attempts to make things a little bit more portable, or at least
       more convenient and native-friendly for non-English users.  The system
       affects character sets and encoding(3,n), and date and time(1,2,n) format-
       ting--amongst other things.

       If you really want to be international, you should consider Unicode.
       See perluniintro and perlunicode for more information.

       If you want to use non-ASCII bytes (outside the bytes 0x00..0x7f) in(1,8)
       the "source code" of your code, to be portable you have to be explicit
       about what bytes they are.  Someone might for example be using your
       code under a UTF-8 locale(3,5,7), in(1,8) which case random(3,4,6) native bytes might be
       illegal ("Malformed UTF-8 ...")  This means that for example embedding
       ISO 8859-1 bytes beyond 0x7f into your strings might cause trouble
       later.  If the bytes are native 8-bit bytes, you can use the "bytes"
       pragma.  If the bytes are in(1,8) a string(3,n) (regular expression being a curi-
       ous string(3,n)), you can often also use the "\xHH" notation instead of
       embedding the bytes as-is.  If they are in(1,8) some particular legacy
       encoding(3,n) (ether single-byte or something more complicated), you can use
       the "encoding(3,n)" pragma.  (If you want to write(1,2) your code in(1,8) UTF-8, you
       can use either the "utf8" pragma, or the "encoding(3,n)" pragma.)  The
       "bytes" and "utf8" pragmata are available since Perl 5.6.0, and the
       "encoding(3,n)" pragma since Perl 5.8.0.

       System Resources

       If your code is destined for systems with severely constrained (or
       missing!) virtual(5,8) memory systems then you want to be especially mindful
       of avoiding wasteful constructs such as:

           # NOTE: this is no longer "bad" in(1,8) perl5.005
           for (0..10000000) {}                       # bad
           for (my $x = 0; $x <= 10000000; ++$x) {}   # good

           @lines = <VERY_LARGE_FILE>;                # bad

           while (<FILE>) {$file(1,n) .= $_}               # sometimes bad
           $file(1,n) = join(1,n)('', <FILE>);                  # better

       The last two constructs may appear unintuitive to most people.  The
       first repeatedly grows a string(3,n), whereas the second allocates a large
       chunk of memory in(1,8) one go.  On some systems, the second is more effi-
       cient that the first.


       Most multi-user platforms provide basic levels of security, usually
       implemented at the filesystem level.  Some, however, do not-- unfortu-
       nately.  Thus the notion of user id, or "home" directory, or even the
       state of being logged-in, may be unrecognizable on many platforms.  If
       you write(1,2) programs that are security-conscious, it is usually best to
       know what type of system you will be running under so that you can
       write(1,2) code explicitly for that platform (or class of platforms).

       Don't assume the UNIX filesystem access(2,5) semantics: the operating system
       or the filesystem may be using some ACL systems, which are richer lan-
       guages than the usual rwx.  Even if(3,n) the rwx exist, their semantics
       might be different.

       (From security viewpoint testing for permissions before attempting to
       do something is silly anyway: if(3,n) one tries this, there is potential for
       race conditions-- someone or something might change the permissions
       between the permissions check and the actual operation.  Just try the

       Don't assume the UNIX user and group semantics: especially, don't
       expect the $< and $> (or the $( and $)) to work for switching identi-
       ties (or memberships).

       Don't assume set-uid and set-gid semantics. (And even if(3,n) you do, think
       twice: set-uid and set-gid are a known can of security worms.)


       For those times when it is necessary to have platform-specific code,
       consider keeping the platform-specific code in(1,8) one place, making port-
       ing to other platforms easier.  Use the Config module and the special
       variable $^O to differentiate platforms, as described in(1,8) "PLATFORMS".

       Be careful in(1,8) the tests you supply with your module or programs.  Mod-
       ule code may be fully portable, but its tests might not be.  This often
       happens when tests spawn off other processes or call external programs
       to aid in(1,8) the testing, or when (as noted above) the tests assume cer-
       tain things about the filesystem and paths.  Be careful not to depend
       on a specific output style for errors, such as when checking $! after a
       failed system call.  Using $! for anything else than displaying it as
       output is doubtful (though see the Errno module for testing reasonably
       portably for error(8,n) value). Some platforms expect a certain output for-
       mat, and Perl on those platforms may have been adjusted accordingly.
       Most specifically, don't anchor a regex(3,7) when testing an error(8,n) value.

CPAN Testers
       Modules uploaded to CPAN are tested by a variety of volunteers on dif-
       ferent platforms.  These CPAN testers are notified by mail(1,8) of each new
       upload, and reply to the list with PASS, FAIL, NA (not applicable to
       this platform), or UNKNOWN (unknown), along with any relevant nota-

       The purpose of the testing is twofold: one, to help developers fix any
       problems in(1,8) their code that crop up because of lack of testing on other
       platforms; two, to provide users(1,5) with information about whether a given
       module works on a given platform.

       Mailing list:
       Testing results:

       As of version(1,3,5) 5.002, Perl is built with a $^O variable that indicates
       the operating system it was built on.  This was implemented to help
       speed up code that would otherwise have to "use Config" and use the
       value of $Config{osname}.  Of course, to get more detailed information
       about the system, looking into %Config is certainly recommended.

       %Config cannot always be trusted, however, because it was built at com-
       pile time.  If perl was built in(1,8) one place, then transferred elsewhere,
       some values may be wrong.  The values may even have been edited after
       the fact.


       Perl works on a bewildering variety of Unix and Unix-like platforms
       (see e.g. most of the files in(1,8) the hints/ directory in(1,8) the source code
       kit).  On most of these systems, the value of $^O (hence $Con-
       fig{'osname'}, too) is determined either by lowercasing and stripping
       punctuation from the first field of the string(3,n) returned by typing
       "uname(1,2) -a" (or a similar command) at the shell prompt or by testing the
       file(1,n) system for the presence of uniquely named(5,8) files such as a kernel
       or header file.  Here, for example, are a few of the more popular Unix

           uname(1,2)         $^O        $Config{'archname'}
           AIX           aix        aix
           BSD/OS        bsdos      i386-bsdos
           Darwin        darwin     darwin
           dgux          dgux       AViiON-dgux
           DYNIX/ptx     dynixptx   i386-dynixptx
           FreeBSD       freebsd    freebsd-i386
           Linux         linux      arm-linux
           Linux         linux      i386-linux
           Linux         linux      i586-linux
           Linux         linux      ppc-linux
           HP-UX         hpux       PA-RISC1.1
           IRIX          irix       irix
           Mac OS X      darwin     darwin
           MachTen PPC   machten    powerpc-machten
           NeXT 3        next       next-fat
           NeXT 4        next       OPENSTEP-Mach
           openbsd       openbsd    i386-openbsd
           OSF1          dec_osf    alpha-dec_osf
           reliantunix-n svr4       RM400-svr4
           SCO_SV        sco_sv     i386-sco_sv
           SINIX-N       svr4       RM400-svr4
           sn4609        unicos     CRAY_C90-unicos
           sn6521        unicosmk   t3e-unicosmk
           sn9617        unicos     CRAY_J90-unicos
           SunOS         solaris    sun4-solaris
           SunOS         solaris    i86pc-solaris
           SunOS4        sunos      sun4-sunos

       Because the value of $Config{archname} may depend on the hardware
       architecture, it can vary more than the value of $^O.

       DOS and Derivatives

       Perl has long been ported to Intel-style microcomputers running under
       systems like PC-DOS, MS-DOS, OS/2, and most Windows platforms you can
       bring yourself to mention (except for Windows CE, if(3,n) you count that).
       Users familiar with COMMAND.COM or CMD.EXE style shells should be aware
       that each of these file(1,n) specifications may have subtle differences:

           $filespec0 = "c:/foo/bar/file.txt";
           $filespec1 = "c:\\foo\\bar\\file.txt";
           $filespec2 = 'c:\foo\bar\file.txt';
           $filespec3 = 'c:\\foo\\bar\\file.txt';

       System calls accept(2,8) either "/" or "\" as the path separator.  However,
       many command-line utilities of DOS vintage treat "/" as the option pre-
       fix, so may get confused by filenames containing "/".  Aside from call-
       ing any external programs, "/" will work just fine, and probably bet-
       ter, as it is more consistent with popular usage, and avoids the prob-
       lem of remembering what to backwhack and what not to.

       The DOS FAT filesystem can accommodate only "8.3" style filenames.
       Under the "case-insensitive, but case-preserving" HPFS (OS/2) and NTFS
       (NT) filesystems you may have to be careful about case returned with
       functions like "readdir(2,3)" or used with functions like "open(2,3,n)" or

       DOS also treats several filenames as special, such as AUX, PRN, NUL,
       CON, COM1, LPT1, LPT2, etc.  Unfortunately, sometimes these filenames
       won't even work if(3,n) you include an explicit directory prefix.  It is
       best to avoid such filenames, if(3,n) you want your code to be portable to
       DOS and its derivatives.  It's hard to know what these all are, unfor-

       Users of these operating systems may also wish to make use of scripts
       such as pl2bat.bat or pl2cmd to put wrappers around your scripts.

       Newline ("\n") is translated as "\015\012" by STDIO when reading from
       and writing to files (see "Newlines").  "binmode(FILEHANDLE)" will keep
       "\n" translated as "\012" for that filehandle.  Since it is a no-op on
       other systems, "binmode" should be used for cross-platform code that
       deals with binary data.  That's assuming you realize in(1,8) advance that
       your data is in(1,8) binary.  General-purpose programs should often assume
       nothing about their data.

       The $^O variable and the $Config{archname} values for various DOSish
       perls are as follows:

            OS            $^O      $Config{archname}   ID    Version
            MS-DOS        dos        ?
            PC-DOS        dos        ?
            OS/2          os2        ?
            Windows 3.1   ?          ?                 0      3 01
            Windows 95    MSWin32    MSWin32-x86       1      4 00
            Windows 98    MSWin32    MSWin32-x86       1      4 10
            Windows ME    MSWin32    MSWin32-x86       1      ?
            Windows NT    MSWin32    MSWin32-x86       2      4 xx
            Windows NT    MSWin32    MSWin32-ALPHA     2      4 xx
            Windows NT    MSWin32    MSWin32-ppc       2      4 xx
            Windows 2000  MSWin32    MSWin32-x86       2      5 xx
            Windows XP    MSWin32    MSWin32-x86       2      ?
            Windows CE    MSWin32    ?                 3
            Cygwin        cygwin     ?

       The various MSWin32 Perl's can distinguish the OS they are running on
       via the value of the fifth element of the list returned from
       Win32::GetOSVersion().  For example:

           if(3,n) ($^O eq 'MSWin32') {
               my @os_version_info = Win32::GetOSVersion();
               print +('3.1','95','NT')[$os_version_info[4]],"\n";

       There are also Win32::IsWinNT() and Win32::IsWin95(), try "perldoc
       Win32", and as of libwin32 0.19 (not part of the core Perl distribu-
       tion) Win32::GetOSName().  The very portable POSIX::uname() will work

           c:\> perl -MPOSIX -we "print join(1,n) '|', uname(1,2)"
           Windows NT|moonru|5.0|Build 2195 (Service Pack 2)|x86

       Also see:

          The djgpp environment for DOS, and

          The EMX environment for DOS, OS/2, etc.,
   Also perlos2.

          Build instructions for Win32 in(1,8) perlwin32, or under the Cygnus
           environment in(1,8) perlcygwin.

          The "Win32::*" modules in(1,8) Win32.

          The ActiveState Pages,

          The Cygwin environment for Win32; README.cygwin (installed as perl-

          The U/WIN environment for Win32,

          Build instructions for OS/2, perlos2

       Mac OS

       Any module requiring XS compilation is right out for most people,
       because MacPerl is built using non-free (and non-cheap!) compilers.
       Some XS modules that can work with MacPerl are built and distributed in(1,8)
       binary form on CPAN.

       Directories are specified as:

           volume:folder:file              for absolute pathnames
           volume:folder:                  for absolute pathnames
           :folder:file                    for relative pathnames
           :folder:                        for relative pathnames
           :file                           for relative pathnames
           file(1,n)                            for relative pathnames

       Files are stored in(1,8) the directory in(1,8) alphabetical order.  Filenames are
       limited to 31 characters, and may include any character except for null
       and ":", which is reserved as the path separator.

       Instead of "flock(1,2)", see "FSpSetFLock" and "FSpRstFLock" in(1,8) the
       Mac::Files module, or "chmod(1,2)(0444, ...)" and "chmod(1,2)(0666, ...)".

       In the MacPerl application, you can't run a program from the command
       line; programs that expect @ARGV to be populated can be edited with
       something like the following, which brings up a dialog(1,3,n) box asking for
       the command line arguments.

           if(3,n) (!@ARGV) {
               @ARGV = split(1,n) /\s+/, MacPerl::Ask('Arguments?');

       A MacPerl script saved as a "droplet" will populate @ARGV with the full
       pathnames of the files dropped onto the script.

       Mac users(1,5) can run programs under a type of command line interface under
       MPW (Macintosh Programmer's Workshop, a free development environment
       from Apple).  MacPerl was first introduced as an MPW tool, and MPW can
       be used like a shell:

           perl myscript.plx some arguments

       ToolServer is another app from Apple that provides access(2,5) to MPW tools
       from MPW and the MacPerl app, which allows MacPerl programs to use
       "system", backticks, and piped "open(2,3,n)".

       "Mac OS" is the proper name for the operating system, but the value in(1,8)
       $^O is "MacOS".  To determine architecture, version(1,3,5), or whether the
       application or MPW tool version(1,3,5) is running, check:

           $is_app    = $MacPerl::Version =~ /App/;
           $is_tool   = $MacPerl::Version =~ /MPW/;
           ($version(1,3,5)) = $MacPerl::Version =~ /^(\S+)/;
           $is_ppc    = $MacPerl::Architecture eq 'MacPPC';
           $is_68k    = $MacPerl::Architecture eq 'Mac68K';

       Mac OS X, based on NeXT's OpenStep OS, runs MacPerl natively, under the
       "Classic" environment.  There is no "Carbon" version(1,3,5) of MacPerl to run
       under the primary Mac OS X environment.  Mac OS X and its Open Source
       version(1,3,5), Darwin, both run Unix perl natively.

       Also see:

          MacPerl Development, .

          The MacPerl Pages, .

          The MacPerl mailing lists, .


       Perl on VMS is discussed in(1,8) perlvms in(1,8) the perl distribution.  Perl on
       VMS can accept(2,8) either VMS- or Unix-style file(1,n) specifications as in(1,8)
       either of the following:

           $ perl -ne "print if(3,n) /perl_setup/i" SYS$LOGIN:LOGIN.COM
           $ perl -ne "print if(3,n) /perl_setup/i" /sys$login(1,3,5)/

       but not a mixture of both as in:

           $ perl -ne "print if(3,n) /perl_setup/i" sys$login:/
           Can't open(2,3,n) sys$login:/ file(1,n) specification syntax error(8,n)

       Interacting with Perl from the Digital Command Language (DCL) shell
       often requires a different set(7,n,1 builtins) of quotation marks than Unix shells do.
       For example:

           $ perl -e "print ""Hello, world.\n"""
           Hello, world.

       There are several ways to wrap your perl scripts in(1,8) DCL .COM files, if(3,n)
       you are so inclined.  For example:

           $ write(1,2) sys$output "Hello from DCL!"
           $ if(3,n) p1 .eqs. ""
           $ then perl -x 'f$environment("PROCEDURE")
           $ else perl -x - 'p1 'p2 'p3 'p4 'p5 'p6 'p7 'p8
           $ deck/dollars="__END__"

           print "Hello from Perl!\n";

           $ endif

       Do take care with "$ ASSIGN/nolog/user SYS$COMMAND: SYS$INPUT" if(3,n) your
       perl-in-DCL script expects to do things like "$read(2,n,1 builtins) = <STDIN>;".

       Filenames are in(1,8) the format "name.extension;version(1,3,5)".  The maximum
       length for filenames is 39 characters, and the maximum length for
       extensions is also 39 characters.  Version is a number from 1 to 32767.
       Valid characters are "/[A-Z0-9$_-]/".

       VMS's RMS filesystem is case-insensitive and does not preserve case.
       "readdir(2,3)" returns lowercased filenames, but specifying a file(1,n) for open-
       ing remains case-insensitive.  Files without extensions have a trailing
       period on them, so doing a "readdir(2,3)" with a file(1,n) named(5,8) A.;5 will return
       a. (though that file(1,n) could be opened with "open(2,3,n)(FH, 'A')").

       RMS had an eight level limit on directory depths from any rooted logi-
       cal (allowing 16 levels overall) prior to VMS 7.2.  Hence
       "PERL_ROOT:[LIB.]" is a valid directory specification but
       "PERL_ROOT:[LIB.]" is not.  Makefile.PL authors might
       have to take this into account, but at least they can refer to the for-
       mer as "/PERL_ROOT/lib/2/3/4/5/6/7/8/".

       The VMS::Filespec module, which gets(3,n) installed as part of the build
       process on VMS, is a pure Perl module that can easily be installed on
       non-VMS platforms and can be helpful for conversions to and from RMS
       native formats.

       What "\n" represents depends on the type of file(1,n) opened.  It usually
       represents "\012" but it could also be "\015", "\012", "\015\012",
       "\000", "\040", or nothing depending on the file(1,n) organiztion and record
       format.  The VMS::Stdio module provides access(2,5) to the special fopen()
       requirements of files with unusual attributes on VMS.

       TCP/IP stacks are optional on VMS, so socket(2,7,n) routines might not be
       implemented.  UDP sockets may not be supported.

       The value of $^O on OpenVMS is "VMS".  To determine the architecture
       that you are running on without resorting to loading all of %Config you
       can examine the content of the @INC array like so:

           if(3,n) (grep(/VMS_AXP/, @INC)) {
               print "I'm on Alpha!\n";

           } elsif (grep(/VMS_VAX/, @INC)) {
               print "I'm on VAX!\n";

           } else {
               print "I'm not so sure about where $^O is...\n";

       On VMS, perl determines the UTC offset from the "SYS$TIMEZONE_DIFFEREN-
       TIAL" logical name.  Although the VMS epoch began at 17-NOV-1858
       00:00:00.00, calls to "localtime" are adjusted to count offsets from
       01-JAN-1970 00:00:00.00, just like Unix.

       Also see:

          README.vms (installed as README_vms), perlvms

          vmsperl list,

           (Put the words "subscribe vmsperl" in(1,8) message body.)

          vmsperl on the web,


       Perl on VOS is discussed in(1,8) README.vos in(1,8) the perl distribution
       (installed as perlvos).  Perl on VOS can accept(2,8) either VOS- or Unix-
       style file(1,n) specifications as in(1,8) either of the following:

           C<< $ perl -ne "print if(3,n) /perl_setup/i" >system>notices >>
           C<< $ perl -ne "print if(3,n) /perl_setup/i" /system/notices >>

       or even a mixture of both as in:

           C<< $ perl -ne "print if(3,n) /perl_setup/i" >system/notices >>

       Even though VOS allows the slash character to appear in(1,8) object names,
       because the VOS port of Perl interprets it as a pathname delimiting
       character, VOS files, directories, or links whose names contain a slash
       character cannot be processed.  Such files must be renamed before they
       can be processed by Perl.  Note that VOS limits file(1,n) names to 32 or
       fewer characters.

       Perl on VOS can be built using two different compilers and two differ-
       ent versions of the POSIX runtime.  The recommended method for building
       full Perl is with the GNU C compiler and the generally-available ver-
       sion(1,3,5) of VOS POSIX support.  See README.vos (installed as perlvos) for
       restrictions that apply when Perl is built using the VOS Standard C
       compiler or the alpha version(1,3,5) of VOS POSIX support.

       The value of $^O on VOS is "VOS".  To determine the architecture that
       you are running on without resorting to loading all of %Config you can
       examine the content of the @INC array like so:

           if(3,n) ($^O =~ /VOS/) {
               print "I'm on a Stratus box!\n";
           } else {
               print "I'm not on a Stratus box!\n";

           if(3,n) (grep(/860/, @INC)) {
               print "This box is a Stratus XA/R!\n";

           } elsif (grep(/7100/, @INC)) {
               print "This box is a Stratus HP 7100 or 8xxx!\n";

           } elsif (grep(/8000/, @INC)) {
               print "This box is a Stratus HP 8xxx!\n";

           } else {
               print "This box is a Stratus 68K!\n";

       Also see:

          README.vos (installed as perlvos)

          The VOS mailing list.

           There is no specific mailing list for Perl on VOS.  You can post
           comments to the comp.sys.stratus newsgroup, or subscribe to the
           general Stratus mailing list.  Send a letter with "subscribe
           Info-Stratus" in(1,8) the message body to

          VOS Perl on the web at http://ftp.stra-

       EBCDIC Platforms

       Recent versions of Perl have been ported to platforms such as OS/400 on
       AS/400 minicomputers as well as OS/390, VM/ESA, and BS2000 for S/390
       Mainframes.  Such computers use EBCDIC character sets internally (usu-
       ally Character Code Set ID 0037 for OS/400 and either 1047 or POSIX-BC
       for S/390 systems).  On the mainframe perl currently works under the
       "Unix system services for OS/390" (formerly known as OpenEdition),
       VM/ESA OpenEdition, or the BS200 POSIX-BC system (BS2000 is supported
       in(1,8) perl 5.6 and greater).  See perlos390 for details.  Note that for
       OS/400 there is also a port of Perl 5.8.1/5.9.0 or later to the PASE
       which is ASCII-based (as opposed to ILE which is EBCDIC-based), see

       As of R2.5 of USS for OS/390 and Version 2.3 of VM/ESA these Unix sub-
       systems do not support the "#!" shebang trick for script invocation.
       Hence, on OS/390 and VM/ESA perl scripts can be executed with a header
       similar to the following simple script:

           : # use perl
               eval 'exec(3,n,1 builtins) /usr/local/bin/perl -S $0 ${1+"$@"}'
                   if(3,n) 0;
           #!/usr/local/bin/perl     # just a comment really

           print "Hello from perl!\n";

       OS/390 will support the "#!" shebang trick in(1,8) release 2.8 and beyond.
       Calls to "system" and backticks can use POSIX shell syntax on all S/390

       On the AS/400, if(3,n) PERL5 is in(1,8) your library list, you may need to wrap
       your perl scripts in(1,8) a CL procedure to invoke them like so:

             CALL PGM(PERL5/PERL) PARM('/QOpenSys/')

       This will invoke the perl script in(1,8) the root of the QOpenSys
       file(1,n) system.  On the AS/400 calls to "system" or backticks must use CL

       On these platforms, bear in(1,8) mind that the EBCDIC character set(7,n,1 builtins) may have
       an effect on what happens with some perl functions (such as "chr",
       "pack(3,n,n pack-old)", "print", "printf(1,3,1 builtins)", "ord", "sort(1,3)", "sprintf", "unpack"), as well
       as bit-fiddling with ASCII constants using operators like "^", "&" and
       "|", not to mention dealing with socket(2,7,n) interfaces to ASCII computers
       (see "Newlines").

       Fortunately, most web servers for the mainframe will correctly trans-
       late the "\n" in(1,8) the following statement to its ASCII equivalent ("\r"
       is the same under both Unix and OS/390 & VM/ESA):

           print "Content-type: text/html\r\n\r\n";

       The values of $^O on some of these platforms includes:

           uname(1,2)         $^O        $Config{'archname'}
           OS/390        os390      os390
           OS400         os400      os400
           POSIX-BC      posix-bc   BS2000-posix-bc
           VM/ESA        vmesa      vmesa

       Some simple tricks for determining if(3,n) you are running on an EBCDIC
       platform could include any of the following (perhaps all):

           if(3,n) ("\t" eq "\05")   { print "EBCDIC may be spoken here!\n"; }

           if(3,n) (ord('A') == 193) { print "EBCDIC may be spoken here!\n"; }

           if(3,n) (chr(169) eq 'z') { print "EBCDIC may be spoken here!\n"; }

       One thing you may not want to rely on is the EBCDIC encoding(3,n) of punctu-
       ation characters since these may differ from code page to code page
       (and once your module or script is rumoured to work with EBCDIC, folks
       will want it to work with all EBCDIC character sets).

       Also see:


           perlos390, README.os390, perlbs2000, README.vmesa, perlebcdic.

          The list is for discussion of porting issues as
           well as general usage issues for all EBCDIC Perls.  Send a message
           body of "subscribe perl-mvs" to

          AS/400 Perl information at as well
           as on CPAN in(1,8) the ports/ directory.

       Acorn RISC OS

       Because Acorns use ASCII with newlines ("\n") in(1,8) text files as "\012"
       like Unix, and because Unix filename emulation is turned on by default,
       most simple scripts will probably work "out of the box".  The native
       filesystem is modular, and individual filesystems are free to be case-
       sensitive or insensitive, and are usually case-preserving.  Some native
       filesystems have name length limits, which file(1,n) and directory names are
       silently truncated to fit.  Scripts should be aware that the standard
       filesystem currently has a name length limit of 10 characters, with up
       to 77 items in(1,8) a directory, but other filesystems may not impose such

       Native filenames are of the form



           Special_Field is not usually present, but may contain . and $ .
           Filesystem =~ m|[A-Za-z0-9_]|
           DsicName   =~ m|[A-Za-z0-9_/]|
           $ represents the root directory
           . is the path separator
           @ is the current directory (per filesystem but machine global)
           ^ is the parent directory
           Directory and File =~ m|[^\0- "\.\$\%\&:\@\\^\|\177]+|

       The default filename translation is roughly "tr|/.|./|;"

       Note that ""ADFS::HardDisk.$.File" ne 'ADFS::HardDisk.$.File'" and that
       the second stage of "$" interpolation in(1,8) regular expressions will fall
       foul of the $. if(3,n) scripts are not careful.

       Logical paths specified by system variables containing comma-separated
       search lists are also allowed; hence "System:Modules" is a valid file-
       name, and the filesystem will prefix "Modules" with each section of
       "System$Path" until a name is made that points to an object on disk.
       Writing to a new file(1,n) "System:Modules" would be allowed only if(3,n) "Sys-
       tem$Path" contains a single item list.  The filesystem will also expand
       system variables in(1,8) filenames if(3,n) enclosed in(1,8) angle brackets, so "<Sys-
       tem$Dir>.Modules" would look(1,8,3 Search::Dict) for the file(1,n) "$ENV{'System$Dir'} . 'Mod-
       ules'".  The obvious implication of this is that fully qualified file-
       names can start with "<>" and should be protected when "open(2,3,n)" is used
       for input.

       Because "." was in(1,8) use as a directory separator and filenames could not
       be assumed to be unique after 10 characters, Acorn implemented the C
       compiler to strip the trailing ".c" ".h" ".s" and ".o" suffix from
       filenames specified in(1,8) source code and store the respective files in(1,8)
       subdirectories named(5,8) after the suffix.  Hence files are translated:

           C:foo.h        (logical path variable)
           sys/os.h        sys.h.os       (C compiler groks Unix-speak)
           10charname.c    c.10charname
           10charname.o    o.10charname
           11charname_.c   c.11charname   (assuming filesystem truncates at 10)

       The Unix emulation library's translation of filenames to native assumes
       that this sort(1,3) of translation is required, and it allows a user-defined
       list of known suffixes that it will transpose in(1,8) this fashion.  This
       may seem transparent, but consider that with these rules
       "foo/bar/baz.h" and "foo/bar/h/baz" both map to "", and
       that "readdir(2,3)" and "glob(1,3,7,n)" cannot and do not attempt to emulate the
       reverse mapping.  Other "."'s in(1,8) filenames are translated to "/".

       As implied above, the environment accessed through %ENV is global, and
       the convention is that program specific environment variables are of
       the form "Program$Name".  Each filesystem maintains a current direc-
       tory, and the current filesystem's current directory is the global cur-
       rent directory.  Consequently, sociable programs don't change the cur-
       rent directory but rely on full pathnames, and programs (and Makefiles)
       cannot assume that they can spawn a child process which can change the
       current directory without affecting its parent (and everyone else for
       that matter).

       Because native operating system filehandles are global and are cur-
       rently allocated down from 255, with 0 being a reserved value, the Unix
       emulation library emulates Unix filehandles.  Consequently, you can't
       rely on passing "STDIN", "STDOUT", or "STDERR" to your children.

       The desire of users(1,5) to express filenames of the form "<Foo$Dir>.Bar" on
       the command line unquoted causes problems, too: `` command output cap-
       ture has to perform a guessing game.  It assumes that a string(3,n)
       "<[^<>]+\$[^<>]>" is a reference to an environment variable, whereas
       anything else involving "<" or ">" is redirection, and generally man-
       ages to be 99% right.  Of course, the problem remains that scripts can-
       not rely on any Unix tools being available, or that any tools found
       have Unix-like command line arguments.

       Extensions and XS are, in(1,8) theory, buildable by anyone using free tools.
       In practice, many don't, as users(1,5) of the Acorn platform are used to
       binary distributions.  MakeMaker does run, but no available make cur-
       rently copes with MakeMaker's makefiles; even if(3,n) and when this should
       be fixed, the lack of a Unix-like shell will cause problems with make-
       file(1,n) rules, especially lines of the form "cd sdbm && make all", and
       anything using quoting.

       "RISC OS" is the proper name for the operating system, but the value in(1,8)
       $^O is "riscos" (because we don't like shouting).

       Other perls

       Perl has been ported to many platforms that do not fit into any of the
       categories listed above.  Some, such as AmigaOS, Atari MiNT, BeOS, HP
       MPE/iX, QNX, Plan 9, and VOS, have been well-integrated into the stan-
       dard Perl source code kit.  You may need to see the ports/ directory on
       CPAN for information, and possibly binaries, for the likes of: aos,
       Atari ST, lynxos, riscos, Novell Netware, Tandem Guardian, etc.  (Yes,
       we know that some of these OSes may fall under the Unix category, but
       we are not a standards body.)

       Some approximate operating system names and their $^O values in(1,8) the
       "OTHER" category include:

           OS            $^O        $Config{'archname'}
           Amiga DOS     amigaos    m68k-amigos
           BeOS          beos
           MPE/iX        mpeix      PA-RISC1.1

       See also:

          Amiga, README.amiga (installed as perlamiga).

          Atari, and Guido Flohr's web page

          Be OS, README.beos

          HP 300 MPE/iX, README.mpeix and Mark Bixby's web page

          A free perl5-based PERL.NLM for Novell Netware is available in(1,8) pre-
           compiled binary and source code form from as
           well as from CPAN.

          Plan 9, README.plan9

       Listed below are functions that are either completely unimplemented or
       else have been implemented differently on various platforms.  Following
       each description will be, in(1,8) parentheses, a list of platforms that the
       description applies to.

       The list may well be incomplete, or even wrong in(1,8) some places.  When in(1,8)
       doubt, consult the platform-specific README files in(1,8) the Perl source
       distribution, and any other documentation resources accompanying a
       given port.

       Be aware, moreover, that even among Unix-ish systems there are varia-

       For many functions, you can also query %Config, exported by default
       from the Config module.  For example, to check whether the platform has
       the "lstat" call, check $Config{d_lstat}.  See Config for a full
       description of available variables.

       Alphabetical Listing of Perl Functions

       -X EXPR
       -X      "-r", "-w", and "-x" have a limited meaning only; directories
               and applications are executable, and there are no uid/gid con-
               siderations.  "-o" is not supported.  (Mac OS)

               "-r", "-w", "-x", and "-o" tell whether the file(1,n) is accessible,
               which may not reflect UIC-based file(1,n) protections.  (VMS)

               "-s" returns the size of the data fork, not the total size of
               data fork plus resource fork.  (Mac OS).

               "-s" by name on an open(2,3,n) file(1,n) will return the space reserved on
               disk, rather than the current extent.  "-s" on an open(2,3,n) filehan-
               dle returns the current size.  (RISC OS)

               "-R", "-W", "-X", "-O" are indistinguishable from "-r", "-w",
               "-x", "-o". (Mac OS, Win32, VMS, RISC OS)

               "-b", "-c", "-k", "-g", "-p", "-u", "-A" are not implemented.
               (Mac OS)

               "-g", "-k", "-l", "-p", "-u", "-A" are not particularly mean-
               ingful.  (Win32, VMS, RISC OS)

               "-d" is true if(3,n) passed a device spec without an explicit direc-
               tory.  (VMS)

               "-T" and "-B" are implemented, but might misclassify Mac text
               files with foreign characters; this is the case will all plat-
               forms, but may affect Mac OS often.  (Mac OS)

               "-x" (or "-X") determine if(3,n) a file(1,n) ends in(1,8) one of the exe-
               cutable suffixes.  "-S" is meaningless.  (Win32)

               "-x" (or "-X") determine if(3,n) a file(1,n) has an executable file(1,n) type.
               (RISC OS)

       binmode FILEHANDLE
               Meaningless.  (Mac OS, RISC OS)

               Reopens file(1,n) and restores pointer; if(3,n) function fails, underly-
               ing filehandle may be closed, or pointer may be in(1,8) a different
               position.  (VMS)

               The value returned by "tell" may be affected after the call,
               and the filehandle may be flushed. (Win32)

       chmod(1,2) LIST
               Only limited meaning.  Disabling/enabling write(1,2) permission is
               mapped to locking/unlocking the file. (Mac OS)

               Only good for changing "owner" read-write access(2,5), "group", and
               "other" bits are meaningless. (Win32)

               Only good for changing "owner" and "other" read-write access.
               (RISC OS)

               Access permissions are mapped onto VOS access-control list
               changes. (VOS)

               The actual permissions set(7,n,1 builtins) depend on the value of the "CYGWIN"
               in(1,8) the SYSTEM environment settings.  (Cygwin)

       chown(1,2) LIST
               Not implemented. (Mac OS, Win32, Plan 9, RISC OS, VOS)

               Does nothing, but won't fail. (Win32)

       chroot(1,2) FILENAME
       chroot(1,2)  Not implemented. (Mac OS, Win32, VMS, Plan 9, RISC OS, VOS,

       crypt PLAINTEXT,SALT
               May not be available if(3,n) library or source was not provided when
               building perl. (Win32)

               Not implemented. (VOS)

       dbmclose HASH
               Not implemented. (VMS, Plan 9, VOS)

       dbmopen HASH,DBNAME,MODE
               Not implemented. (VMS, Plan 9, VOS)

       dump LABEL
               Not useful. (Mac OS, RISC OS)

               Not implemented. (Win32)

               Invokes VMS debugger. (VMS)

       exec(3,n,1 builtins) LIST
               Not implemented. (Mac OS)

               Implemented via Spawn. (VM/ESA)

               Does not automatically flush(8,n) output handles on some platforms.
               (SunOS, Solaris, HP-UX)

       exit(3,n,1 builtins) EXPR
       exit(3,n,1 builtins)    Emulates UNIX exit(3,n,1 builtins)() (which considers "exit(3,n,1 builtins) 1" to indicate an
               error(8,n)) by mapping the 1 to SS$_ABORT (44).  This behavior may
               be overridden with the pragma "use vmsish 'exit(3,n,1 builtins)'".  As with the
               CRTL's exit(3,n,1 builtins)() function, "exit(3,n,1 builtins) 0" is also mapped to an exit(3,n,1 builtins) sta-
               tus of SS$_NORMAL (1); this mapping cannot be overridden.  Any
               other argument to exit(3,n,1 builtins)() is used directly as Perl's exit(3,n,1 builtins) sta-
               tus. (VMS)

               Not implemented. (Win32, VMS)

       flock(1,2) FILEHANDLE,OPERATION
               Not implemented (Mac OS, VMS, RISC OS, VOS).

               Available only on Windows NT (not on Windows 95). (Win32)

       fork    Not implemented. (Mac OS, AmigaOS, RISC OS, VOS, VM/ESA, VMS)

               Emulated using multiple interpreters.  See perlfork.  (Win32)

               Does not automatically flush(8,n) output handles on some platforms.
               (SunOS, Solaris, HP-UX)

               Not implemented. (Mac OS, RISC OS)

       getpgrp PID
               Not implemented. (Mac OS, Win32, VMS, RISC OS, VOS)

       getppid Not implemented. (Mac OS, Win32, RISC OS)

       getpriority WHICH,WHO
               Not implemented. (Mac OS, Win32, VMS, RISC OS, VOS, VM/ESA)

       getpwnam NAME
               Not implemented. (Mac OS, Win32)

               Not useful. (RISC OS)

       getgrnam NAME
               Not implemented. (Mac OS, Win32, VMS, RISC OS)

       getnetbyname NAME
               Not implemented. (Mac OS, Win32, Plan 9)

       getpwuid UID
               Not implemented. (Mac OS, Win32)

               Not useful. (RISC OS)

       getgrgid GID
               Not implemented. (Mac OS, Win32, VMS, RISC OS)

       getnetbyaddr ADDR,ADDRTYPE
               Not implemented. (Mac OS, Win32, Plan 9)

       getprotobynumber NUMBER
               Not implemented. (Mac OS)

       getservbyport PORT,PROTO
               Not implemented. (Mac OS)

               Not implemented. (Mac OS, Win32, VM/ESA)

               Not implemented. (Mac OS, Win32, VMS, VM/ESA)

               "gethostbyname('localhost')" does not work everywhere: you may
               have to use "gethostbyname('')". (Mac OS, Irix 5)

               Not implemented. (Mac OS, Win32)

               Not implemented. (Mac OS, Win32, Plan 9)

               Not implemented. (Mac OS, Win32, Plan 9)

               Not implemented. (Win32, Plan 9)

       sethostent STAYOPEN
               Not implemented. (Mac OS, Win32, Plan 9, RISC OS)

       setnetent STAYOPEN
               Not implemented. (Mac OS, Win32, Plan 9, RISC OS)

       setprotoent STAYOPEN
               Not implemented. (Mac OS, Win32, Plan 9, RISC OS)

       setservent STAYOPEN
               Not implemented. (Plan 9, Win32, RISC OS)

               Not implemented. (Mac OS, MPE/iX, VM/ESA, Win32)

               Not implemented. (Mac OS, MPE/iX, RISC OS, VM/ESA, VMS, Win32)

               Not implemented. (Mac OS, Win32)

               Not implemented. (Mac OS, Win32, Plan 9)

               Not implemented. (Mac OS, Win32, Plan 9)

               Not implemented. (Plan 9, Win32)

       getsockopt SOCKET,LEVEL,OPTNAME
               Not implemented. (Plan 9)

       glob(1,3,7,n) EXPR
       glob(1,3,7,n)    This operator is implemented via the File::Glob extension on
               most platforms.  See File::Glob for portability information.

               Not implemented. (VMS)

               Available only for socket(2,7,n) handles, and it does what the ioctl-
               socket(2,7,n)() call in(1,8) the Winsock API does. (Win32)

               Available only for socket(2,7,n) handles. (RISC OS)

       kill(1,2,1 builtins) SIGNAL, LIST
               "kill(1,2,1 builtins)(0, LIST)" is implemented for the sake of taint checking;
               use with other signals is unimplemented. (Mac OS)

               Not implemented, hence not useful for taint checking. (RISC OS)

               "kill(1,2,1 builtins)()" doesn't have the semantics of "raise(3,n)()", i.e. it
               doesn't send(2,n) a signal(2,7) to the identified process like it does on
               Unix platforms.  Instead "kill(1,2,1 builtins)($sig, $pid)" terminates the
               process identified by $pid, and makes it exit(3,n,1 builtins) immediately with
               exit(3,n,1 builtins) status $sig.  As in(1,8) Unix, if(3,n) $sig is 0 and the specified
               process exists, it returns true without actually terminating
               it. (Win32)

       link(1,2) OLDFILE,NEWFILE
               Not implemented. (Mac OS, MPE/iX, VMS, RISC OS)

               Link count not updated because hard links are not quite that
               hard (They are sort(1,3) of half-way between hard and soft links).

               Hard links are implemented on Win32 (Windows NT and Windows
               2000) under NTFS only.

       lstat FILEHANDLE
       lstat EXPR
       lstat   Not implemented. (VMS, RISC OS)

               Return values (especially for device and inode) may be bogus.

       msgctl ID,CMD,ARG
       msgget KEY,FLAGS
       msgsnd ID,MSG,FLAGS
       msgrcv ID,VAR,SIZE,TYPE,FLAGS
               Not implemented. (Mac OS, Win32, VMS, Plan 9, RISC OS, VOS)

       open(2,3,n) FILEHANDLE,EXPR
       open(2,3,n) FILEHANDLE
               The "|" variants are supported only if(3,n) ToolServer is installed.
               (Mac OS)

               open(2,3,n) to "|-" and "-|" are unsupported. (Mac OS, Win32, RISC OS)

               Opening a process does not automatically flush(8,n) output handles
               on some platforms.  (SunOS, Solaris, HP-UX)

               Very limited functionality. (MiNT)

       readlink(1,2) EXPR
               Not implemented. (Win32, VMS, RISC OS)

       select(2,7,2 select_tut) RBITS,WBITS,EBITS,TIMEOUT
               Only implemented on sockets. (Win32, VMS)

               Only reliable on sockets. (RISC OS)

               Note that the "select(2,7,2 select_tut) FILEHANDLE" form is generally portable.

       semctl ID,SEMNUM,CMD,ARG
       semget KEY,NSEMS,FLAGS
       semop KEY,OPSTRING
               Not implemented. (Mac OS, Win32, VMS, RISC OS, VOS)

               Not implemented. (Mac OS, MPE/iX, VMS, Win32, RISC OS)

       setpgrp PID,PGRP
               Not implemented. (Mac OS, Win32, VMS, RISC OS, VOS)

       setpriority WHICH,WHO,PRIORITY
               Not implemented. (Mac OS, Win32, VMS, RISC OS, VOS)

               Not implemented. (Mac OS, MPE/iX, Win32, RISC OS)

               Not implemented. (Plan 9)

       shmctl ID,CMD,ARG
       shmget KEY,SIZE,FLAGS
       shmread ID,VAR,POS,SIZE
       shmwrite ID,STRING,POS,SIZE
               Not implemented. (Mac OS, Win32, VMS, RISC OS, VOS)

       sockatmark SOCKET
               A relatively recent addition to socket(2,7,n) functions, may not be
               implemented even in(1,8) UNIX platforms.

               Not implemented. (Win32, VMS, RISC OS, VOS, VM/ESA)

       stat(1,2) FILEHANDLE
       stat(1,2) EXPR
       stat(1,2)    Platforms that do not have rdev, blksize, or blocks will return
               these as '', so numeric comparison or manipulation of these
               fields may cause 'not numeric' warnings.

               mtime and atime are the same thing, and ctime is creation time(1,2,n)
               instead of inode change time. (Mac OS).

               ctime not supported on UFS (Mac OS X).

               ctime is creation time(1,2,n) instead of inode change time(1,2,n)  (Win32).

               device and inode are not meaningful.  (Win32)

               device and inode are not necessarily reliable.  (VMS)

               mtime, atime and ctime all return the last modification time.
               Device and inode are not necessarily reliable.  (RISC OS)

               dev, rdev, blksize, and blocks are not available.  inode is not
               meaningful and will differ between stat(1,2) calls on the same file.

               some versions of cygwin when doing a stat(1,2)("foo") and if(3,n) not
               finding it may then attempt to stat(1,2)("foo.exe") (Cygwin)

       symlink OLDFILE,NEWFILE
               Not implemented. (Win32, VMS, RISC OS)

       syscall LIST
               Not implemented. (Mac OS, Win32, VMS, RISC OS, VOS, VM/ESA)

               The traditional "0", "1", and "2" MODEs are implemented with
               different numeric values on some systems.  The flags exported
               by "Fcntl" (O_RDONLY, O_WRONLY, O_RDWR) should work everywhere
               though.  (Mac OS, OS/390, VM/ESA)

       system LIST
               In general, do not assume the UNIX/POSIX semantics that you can
               shift $? right by eight to get the exit(3,n,1 builtins) value, or that "$? &
               127" would give you the number of the signal(2,7) that terminated
               the program, or that "$? & 128" would test true if(3,n) the program
               was terminated by a coredump.  Instead, use the POSIX W*()
               interfaces: for example, use WIFEXITED($?) and WEXITVALUE($?)
               to test for a normal exit(3,n,1 builtins) and the exit(3,n,1 builtins) value, WIFSIGNALED($?)
               and WTERMSIG($?) for a signal(2,7) exit(3,n,1 builtins) and the signal.  Core dump-
               ing is not a portable concept, so there's no portable way to
               test for that.

               Only implemented if(3,n) ToolServer is installed. (Mac OS)

               As an optimization, may not call the command shell specified in(1,8)
               $ENV{PERL5SHELL}.  "system(1, @args)" spawns an external
               process and immediately returns its process designator, without
               waiting for it to terminate.  Return value may be used subse-
               quently in(1,8) "wait" or "waitpid".  Failure to spawn() a subpro-
               cess is indicated by setting $? to "255 << 8".  $? is set(7,n,1 builtins) in(1,8) a
               way compatible with Unix (i.e. the exitstatus of the subprocess
               is obtained by "$? >> 8", as described in(1,8) the documentation).

               There is no shell to process metacharacters, and the native
               standard is to pass a command line terminated by "\n" "\r" or
               "\0" to the spawned program.  Redirection such as "> foo" is
               performed (if(3,n) at all) by the run time(1,2,n) library of the spawned
               program.  "system" list will call the Unix emulation library's
               "exec(3,n,1 builtins)" emulation, which attempts to provide emulation of the
               stdin, stdout, stderr in(1,8) force in(1,8) the parent, providing the
               child program uses a compatible version(1,3,5) of the emulation
               library.  scalar will call the native command line direct and
               no such emulation of a child Unix program will exists.  Mileage
               will vary.  (RISC OS)

               Far from being POSIX compliant.  Because there may be no under-
               lying /bin/sh tries to work around the problem by forking and
               execing the first token in(1,8) its argument string.  Handles basic
               redirection ("<" or ">") on its own behalf. (MiNT)

               Does not automatically flush(8,n) output handles on some platforms.
               (SunOS, Solaris, HP-UX)

               The return value is POSIX-like (shifted up by 8 bits), which
               only allows room for a made-up value derived from the severity
               bits of the native 32-bit condition code (unless overridden by
               "use vmsish 'status'").  For more details see "$?" in(1,8) perlvms.

       times   Only the first entry returned is nonzero. (Mac OS)

               "cumulative" times will be bogus.  On anything other than Win-
               dows NT or Windows 2000, "system" time(1,2,n) will be bogus, and
               "user" time(1,2,n) is actually the time(1,2,n) returned by the clock(3,n)() func-
               tion in(1,8) the C runtime library. (Win32)

               Not useful. (RISC OS)

       truncate(2,7) FILEHANDLE,LENGTH
       truncate(2,7) EXPR,LENGTH
               Not implemented. (Older versions of VMS)

               Truncation to zero-length only. (VOS)

               If a FILEHANDLE is supplied, it must be writable and opened in(1,8)
               append mode (i.e., use "open(2,3,n)(FH, '>>filename')" or
               "sysopen(FH,...,O_APPEND|O_RDWR)".  If a filename is supplied,
               it should not be held open(2,3,n) elsewhere. (Win32)

       umask EXPR
       umask   Returns undef where unavailable, as of version(1,3,5) 5.005.

               "umask" works but the correct permissions are set(7,n,1 builtins) only when the
               file(1,n) is finally closed. (AmigaOS)

       utime LIST
               Only the modification time(1,2,n) is updated. (BeOS, Mac OS, VMS,
               RISC OS)

               May not behave as expected.  Behavior depends on the C runtime
               library's implementation of utime(), and the filesystem being
               used.  The FAT filesystem typically does not support an "access(2,5)
               time(1,2,n)" field, and it may limit timestamps to a granularity of
               two seconds. (Win32)

       waitpid PID,FLAGS
               Not implemented. (Mac OS, VOS)

               Can only be applied to process handles returned for processes
               spawned using "system(1, ...)" or pseudo processes created with
               "fork()". (Win32)

               Not useful. (RISC OS)

       v1.48, 02 February 2001
           Various updates from perl5-porters over the past year, supported
           platforms update(7,n) from Jarkko Hietaniemi.

       v1.47, 22 March 2000
           Various cleanups from Tom Christiansen, including migration of long
           platform listings from perl.

       v1.46, 12 February 2000
           Updates for VOS and MPE/iX. (Peter Prymmer)  Other small changes.

       v1.45, 20 December 1999
           Small changes from 5.005_63 distribution, more changes to EBCDIC

       v1.44, 19 July 1999
           A bunch of updates from Peter Prymmer for $^O values, endianness,
           File::Spec, VMS, BS2000, OS/400.

       v1.43, 24 May 1999
           Added a lot of cleaning up from Tom Christiansen.

       v1.42, 22 May 1999
           Added notes about tests, sprintf/printf(1,3,1 builtins), and epoch offsets.

       v1.41, 19 May 1999
           Lots more little changes to formatting and content.

           Added a bunch of $^O and related values for various platforms;
           fixed mail(1,8) and web addresses, and added and changed miscellaneous
           notes.  (Peter Prymmer)

       v1.40, 11 April 1999
           Miscellaneous changes.

       v1.39, 11 February 1999
           Changes from Jarkko and EMX URL fixes Michael Schwern.  Additional
           note about newlines added.

       v1.38, 31 December 1998
           More changes from Jarkko.

       v1.37, 19 December 1998
           More minor changes.  Merge two separate version(1,3,5) 1.35 documents.

       v1.36, 9 September 1998
           Updated for Stratus VOS.  Also known as version(1,3,5) 1.35.

       v1.35, 13 August 1998
           Integrate more minor changes, plus addition of new sections under
           "ISSUES": "Numbers endianness and Width", "Character sets and char-
           acter encoding(3,n)", "Internationalisation".

       v1.33, 06 August 1998
           Integrate more minor changes.

       v1.32, 05 August 1998
           Integrate more minor changes.

       v1.30, 03 August 1998
           Major update(7,n) for RISC OS, other minor changes.

       v1.23, 10 July 1998
           First public release with perl5.005.

Supported Platforms
       As of September 2003 (the Perl release 5.8.1), the following platforms
       are able to build Perl from the standard source code distribution
       available at

               BSD/OS          (BSDi)
               DOS DJGPP       1)
               EPOC R5
               HI-UXMPP        (Hitachi) (5.8.0 worked but we didn't know it)
               Mac OS Classic
               Mac OS X        (Darwin)
               ReliantUNIX     (formerly SINIX)
               OpenVMS         (formerly VMS)
               Open UNIX       (Unixware) (since Perl 5.8.1/5.9.0)
               OS/400          (using the PASE) (since Perl 5.8.1/5.9.0)
               POSIX-BC        (formerly BS2000)
               SunOS 4
               SUPER-UX        (NEC)
               Tru64 UNIX      (formerly DEC OSF/1, Digital UNIX)
               Win95/98/ME/2K/XP 2)
               z/OS            (formerly OS/390)

               1) in(1,8) DOS mode either the DOS or OS/2 ports can be used
               2) compilers: Borland, MinGW (GCC), VC6

       The following platforms worked with the previous releases (5.6 and
       5.7), but we did not manage either to fix or to test these in(1,8) time(1,2,n) for
       the 5.8.1 release.  There is a very good chance that many of these will
       work fine with the 5.8.1.

               SCO SV
               Windows 3.1

       Known to be broken for 5.8.0 and 5.8.1 (but 5.6.1 and 5.7.2 can be


       The following platforms have been known to build Perl from source in(1,8)
       the past (5.005_03 and earlier), but we haven't been able to verify(1,8)
       their status for the current release, either because the hardware/soft-
       ware platforms are rare or because we don't have an active champion on
       these platforms--or both.  They used to work, though, so go ahead and
       try compiling them, and let of any trouble.

               DDE SMES
               DOS EMX
               MachTen 68k
               Plan 9
               SCO ODT/OSR
               Unisys Dynix

       The following platforms have their own source code distributions and
       binaries available via

                                       Perl release

               OS/400 (ILE)            5.005_02
               Tandem Guardian         5.004

       The following platforms have only binaries available via :

                                       Perl release

               Acorn RISCOS            5.005_02
               AOS                     5.002
               LynxOS                  5.004_02

       Although we do suggest that you always build your own Perl from the
       source code, both for maximal configurability and for security, in(1,8) case
       you are in(1,8) a hurry you can check
       for binary distributions.

       perlaix, perlamiga, perlapollo, perlbeos, perlbs2000, perlce, perlcyg-
       win, perldgux, perldos, perlepoc, perlebcdic, perlfreebsd, perlhurd,
       perlhpux, perlirix, perlmachten, perlmacos, perlmacosx, perlmint,
       perlmpeix, perlnetware, perlos2, perlos390, perlos400, perlplan9, per-
       lqnx, perlsolaris, perltru64, perlunicode, perlvmesa, perlvms, perlvos,
       perlwin32, and Win32.

       Abigail <>, Charles Bailey <>,
       Graham Barr <>, Tom Christiansen <>,
       Nicholas Clark <>, Thomas Dorner <>,
       Andy Dougherty <>, Dominic Dunlop <domo@com->, Neale Ferguson <>, David J. Fiander
       <>, Paul Green <>, M.J.T. Guy
       <>, Jarkko Hietaniemi <>, Luther Huffman
       <>, Nick Ing-Simmons <>,
       Andreas J. Koenig <>, Markus Laker <mlaker@con->, Andrew M. Langmead <>, Larry Moore
       <>, Paul Moore <>,
       Chris Nandor <>, Matthias Neeracher <>,
       Philip Newton <>, Gary Ng <71564.1743@CompuServe.COM>, Tom
       Phoenix <>, Andre Pirard <>,
       Peter Prymmer <>, Hugo van der Sanden
       <>, Gurusamy Sarathy <>, Paul
       J. Schinder <>, Michael G Schwern <schw->, Dan Sugalski <>, Nathan Torkington

perl v5.8.5                       2004-04-23                       PERLPORT(1)

References for this manual (incoming links)