File: perlport.html

package info (click to toggle)
perl-doc-html 5.10.1-1
links: PTS
area: main
in suites: squeeze
size: 33,152 kB
ctags: 24,035
sloc: makefile: 10
file content (2162 lines) | stat: -rw-r--r-- 149,826 bytes
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN"
"http://www.w3.org/TR/html4/loose.dtd">
<html>
<head>
  <title>perlport - perldoc.perl.org</title>
  <meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
  <meta http-equiv="Content-Language" content="en-gb">
  <link href="static/css-20090810.css" rel="stylesheet" rev="stylesheet" type="text/css" media="screen">
</head>

<body onLoad="perldoc.startup();" onPageShow="if (event.persisted) perldoc.startup();">
    <div id="page">
      
      <div id="header">
	<div id="homepage_link">
	  <a href="index.html"></a>
	</div>
        <div id="search_form">
          <form action="search.html" method="GET" id="search">
            <input type="text" name="q" id="search_box" alt="Search">
          </form>
        </div>
      </div>
      
      <div id="body">
        <div id="left_column">
          <div class="side_group">
            <div class="side_panel doc_panel">
              <p>Manual</p>
              <ul>
                <li><a href="index-overview.html">Overview</a>
                <li><a href="index-tutorials.html">Tutorials</a>
                <li><a href="index-faq.html">FAQs</a>
                <li><a href="index-history.html">History / Changes</a>
                <li><a href="index-licence.html">License</a>
              </ul>
            </div>
            <div class="side_panel doc_panel">
              <p>Reference</p>
              <ul>
                <li><a href="index-language.html">Language</a>
                <li><a href="index-functions.html">Functions</a>
                <li><a href="perlop.html">Operators</a>
                <li><a href="perlvar.html">Special Variables</a>
                <li><a href="index-pragmas.html">Pragmas</a>
                <li><a href="index-utilities.html">Utilities</a>
                <li><a href="index-internals.html">Internals</a></li>
                <li><a href="index-platforms.html">Platform Specific</a>
              </ul>
            </div>
            <div class="side_panel doc_panel">
              <p>Modules</p>
              <ul>
		<li>
		
                
                  
		    
		  
		
                  
		    
		  
		
                  
		    
		  
		
                  
		    
		  
		
                  
		    
		  
		
                  
		    
		  
		
                  
		    
		  
		
                  
		    
		  
		
                  
		    
		  
		
                  
		
                  
		
                  
		    
		  
		
                  
		    
		  
		
                  
		    
		  
		
                  
		    
		  
		
                  
		    
		  
		
                  
		
                  
		
                  
		    
		  
		
                  
		    
		  
		
                  
		    
		  
		
                  
		
                  
		
                  
		    
		  
		
                  
		
                  
		
		
                    <a href="index-modules-A.html">A</a>
                    
                      
                        &bull;
                      
                    
                
                    <a href="index-modules-B.html">B</a>
                    
                      
                        &bull;
                      
                    
                
                    <a href="index-modules-C.html">C</a>
                    
                      
                        &bull;
                      
                    
                
                    <a href="index-modules-D.html">D</a>
                    
                      
                        &bull;
                      
                    
                
                    <a href="index-modules-E.html">E</a>
                    
                      
                        <li>
                      
                    
                
                    <a href="index-modules-F.html">F</a>
                    
                      
                        &bull;
                      
                    
                
                    <a href="index-modules-G.html">G</a>
                    
                      
                        &bull;
                      
                    
                
                    <a href="index-modules-H.html">H</a>
                    
                      
                        &bull;
                      
                    
                
                    <a href="index-modules-I.html">I</a>
                    
                      
                        &bull;
                      
                    
                
                    <a href="index-modules-L.html">L</a>
                    
                      
                        <li>
                      
                    
                
                    <a href="index-modules-M.html">M</a>
                    
                      
                        &bull;
                      
                    
                
                    <a href="index-modules-N.html">N</a>
                    
                      
                        &bull;
                      
                    
                
                    <a href="index-modules-O.html">O</a>
                    
                      
                        &bull;
                      
                    
                
                    <a href="index-modules-P.html">P</a>
                    
                      
                        &bull;
                      
                    
                
                    <a href="index-modules-S.html">S</a>
                    
                      
                        <li>
                      
                    
                
                    <a href="index-modules-T.html">T</a>
                    
                      
                        &bull;
                      
                    
                
                    <a href="index-modules-U.html">U</a>
                    
                      
                        &bull;
                      
                    
                
                    <a href="index-modules-X.html">X</a>
                    
                
              </ul>
            </div>
          </div>
          <div class="side_group">
            <div class="side_panel links_panel">
              <p>News / Blogs</p>
              <ul>
        <li><a href="http://perlbuzz.com">Perl Buzz</a></li>
        <li><a href="http://ironman.enlightenedperl.org">Perl Iron Man</a></li>
        <li><a href="http://perlsphere.net">Perlsphere</a></li>
        <li><a href="http://planet.perl.org">Planet Perl</a></li>
	<li><a href="http://news.perlfoundation.org/">TPF news</a></li>
        <li><a href="http://use.perl.org">Use Perl</a></li>
              </ul>
            </div>            
            <div class="side_panel links_panel">
              <p>See also</p>
              <ul>
        <li><a href="http://search.cpan.org">CPAN</a></li>
        <li><a href="http://www.perl.org">Perl.org</a></li>
        <li><a href="http://www.perl.com">Perl.com</a></li>
        <li><a href="http://www.perlfoundation.org/perl5/index.cgi">Perl 5 Wiki</a></li>
        <li><a href="http://jobs.perl.org">Perl Jobs</a></li>
        <li><a href="http://www.pm.org">Perl Mongers</a></li>
        <li><a href="http://www.perlmonks.org">Perl Monks</a></li>
              </ul>
            </div>
            <div class="side_panel links_panel">
              <p>Contact</p>
              <ul>
                <li><a href="http://perl.jonallen.info">Site designed and maintained by Jon&nbsp;Allen&nbsp;(JJ)</a>
                <li><a href="http://perl.jonallen.info">http://perl.jonallen.info</a>
              </ul>
              <ul>
                <li><a href="http://perl.jonallen.info/projects/perldoc">View the project page and FAQs</a>
              </ul>
            </div>
          </div>
        </div>
        <div id="centre_column">
          <div id="content_header">
            <div id="title_bar">
              <div id="page_name">
                <h1>perlport</h1>
              </div>
              <div id="perl_version">
                Perl 5 version 10.1 documentation
              </div>
              <div id="page_links">
                <a href="#" onClick="toolbar.goToTop();return false;">Go to top</a>
		
                  &bull; <a href="#" id="page_index_toggle">Show page index</a>
		
		
              </div>
            </div>
            <div id="breadcrumbs">
                
    <a href="index.html">Home</a> &gt;
    
      
        <a href="index-language.html">Language reference</a> &gt;
      
    
    perlport
  

            </div>
          </div>
          <div id="content_body">
	    <!--[if lt IE 7]>
 <div class="noscript">
   <p>
     <strong>It looks like you're using Internet Explorer 6. This is a very old
     browser which does not offer full support for modern websites.</strong>
   </p>
   <p>
     Unfortunately this means that this website will not work on
     your computer.
   </p>
   <p>
     Don't miss out though! To view the site (and get a better experience from
     many other websites), simply upgrade to
     <a href="http://www.microsoft.com/windows/Internet-explorer/default.aspx">Internet
Explorer 8</a>
     or download an alternative browser such as
     <a href="http://www.mozilla.com/en-US/firefox/firefox.html">Firefox</a>,
     <a href="http://www.apple.com/safari/download/">Safari</a>, or
     <a href="http://www.google.co.uk/chrome">Google Chrome</a>.
   </p>
   <p>
     All of these browsers are free. If you're using a PC at work, you may
     need to contact your IT administrator.
   </p>
 </div>
<![endif]-->
	    <noscript>
	      <div class="noscript">
	      <p>
                <strong>Please note: Many features of this site require JavaScript. You appear to have JavaScript disabled,
	        or are running a non-JavaScript capable web browser.</strong>
	      </p>
	      <p>
		To get the best experience, please enable JavaScript or download a modern web browser such as <a href="http://www.microsoft.com/windows/Internet-explorer/default.aspx">Internet Explorer 8</a>, <a href="http://www.mozilla.com/en-US/firefox/firefox.html">Firefox</a>, <a href="http://www.apple.com/safari/download/">Safari</a>, or <a href="http://www.google.co.uk/chrome">Google Chrome</a>.
              </p>
	      </div>
	    </noscript>
	    <div id="from_search"></div>
            <h1>perlport</h1>


  <!--    -->
<ul><li><a href="#NAME">NAME</a><li><a href="#DESCRIPTION">DESCRIPTION</a><li><a href="#ISSUES">ISSUES</a><ul><li><a href="#Newlines">Newlines</a><li><a href="#Numbers-endianness-and-Width">Numbers endianness and Width</a><li><a href="#Files-and-Filesystems">Files and Filesystems</a><li><a href="#System-Interaction">System Interaction</a><li><a href="#Command-names-versus-file-pathnames">Command names versus file pathnames</a><li><a href="#Networking">Networking</a><li><a href="#Interprocess-Communication-(IPC)">Interprocess Communication (IPC)</a><li><a href="#External-Subroutines-(XS)">External Subroutines (XS)</a><li><a href="#Standard-Modules">Standard Modules</a><li><a href="#Time-and-Date">Time and Date</a><li><a href="#Character-sets-and-character-encoding">Character sets and character encoding</a><li><a href="#Internationalisation">Internationalisation</a><li><a href="#System-Resources">System Resources</a><li><a href="#Security">Security</a><li><a href="#Style">Style</a></ul><li><a href="#CPAN-Testers">CPAN Testers</a><li><a href="#PLATFORMS">PLATFORMS</a><ul><li><a href="#Unix">Unix</a><li><a href="#DOS-and-Derivatives">DOS and Derivatives</a><li><a href="#Mac-OS">Mac OS</a><li><a href="#VMS">VMS</a><li><a href="#VOS">VOS</a><li><a href="#EBCDIC-Platforms">EBCDIC Platforms</a><li><a href="#Acorn-RISC-OS">Acorn RISC OS</a><li><a href="#Other-perls">Other perls</a></ul><li><a href="#FUNCTION-IMPLEMENTATIONS">FUNCTION IMPLEMENTATIONS</a><ul><li><a href="#Alphabetical-Listing-of-Perl-Functions">Alphabetical Listing of Perl Functions</a></ul><li><a href="#Supported-Platforms">Supported Platforms</a><li><a href="#SEE-ALSO">SEE ALSO</a><li><a href="#AUTHORS-%2f-CONTRIBUTORS">AUTHORS / CONTRIBUTORS</a></ul><a name="NAME"></a><h1>NAME</h1>
<p>perlport - Writing portable Perl</p>
<a name="DESCRIPTION"></a><h1>DESCRIPTION</h1>
<p>Perl runs on numerous operating systems.  While most of them share
much in common, they also have their own unique features.</p>
<p>This document is meant to help you to find out what constitutes portable
Perl code.  That way once you make a decision to write portably,
you know where the lines are drawn, and you can stay within them.</p>
<p>There is a tradeoff between taking full advantage of one particular
type of computer and taking advantage of a full range of them.
Naturally, 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 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.  Specifically, 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.</p>
<p>Looking at it another way, writing portable code is usually about
willfully 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.</p>
<p>Be aware of two important points:</p>
<ul>
<li><a name="Not-all-Perl-programs-have-to-be-portable"></a><b>Not all Perl programs have to be portable</b>
<p>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 a given program, then don&#39;t bother.</p>
</li>
<li><a name="Nearly-all-of-Perl-already-_is_-portable"></a><b>Nearly all of Perl already <i>is</i> portable</b>
<p>Don&#39;t be fooled into thinking that it is hard to create portable Perl
code.  It isn&#39;t.  Perl tries its level-best to bridge the gaps between
what&#39;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
writing portable code, and this document is entirely about those issues.</p>
</li>
</ul>
<p>Here&#39;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&#39;t sacrifice much by way of the implementation
choices you can avail yourself of, and at the same time you can give
your users 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.</p>
<p>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
deliberate in your decision.</p>
<p>The material below is separated into three main sections: main issues of
portability (<a href="http://search.cpan.org/perldoc/&quot;ISSUES&quot;">ISSUES</a>), platform-specific issues (<a href="http://search.cpan.org/perldoc/&quot;PLATFORMS&quot;">PLATFORMS</a>), and
built-in perl functions that behave differently on various ports
(<a href="#FUNCTION-IMPLEMENTATIONS">"FUNCTION IMPLEMENTATIONS"</a>).</p>
<p>This information should not be considered complete; it includes possibly
transient information about idiosyncrasies of some of the ports, almost
all of which are in a state of constant evolution.  Thus, this material
should be considered a perpetual work in progress
(<code class="inline"><span class="q">&lt;IMG SRC=&quot;yellow_sign.gif&quot; ALT=&quot;Under Construction&quot;&gt;</span></code>
).</p>
<a name="ISSUES"></a><h1>ISSUES</h1>
<a name="Newlines"></a><h2>Newlines</h2>
<p>In most operating systems, lines in files are terminated by newlines.
Just what is used as a newline may vary from OS to OS.  Unix
traditionally uses <code class="inline">\<span class="n">012</span></code>
, one type of DOSish I/O uses <code class="inline">\<span class="n">015</span>\<span class="n">012</span></code>
,
and Mac OS uses <code class="inline">\<span class="n">015</span></code>
.</p>
<p>Perl uses <code class="inline">\<span class="w">n</span></code>
 to represent the &quot;logical&quot; newline, where what is
logical may depend on the platform in use.  In MacPerl, <code class="inline">\<span class="w">n</span></code>
 always
means <code class="inline">\<span class="n">015</span></code>
.  In DOSish perls, <code class="inline">\<span class="w">n</span></code>
 usually means <code class="inline">\<span class="n">012</span></code>
, but
when accessing a file in &quot;text&quot; mode, STDIO translates it to (or
from) <code class="inline">\<span class="n">015</span>\<span class="n">012</span></code>
, depending on whether you&#39;re reading or writing.
Unix does the same thing on ttys in canonical mode.  <code class="inline">\<span class="n">015</span>\<span class="n">012</span></code>

is commonly referred to as CRLF.</p>
<p>To trim trailing newlines from text lines use chomp().  With default 
settings that function looks for a trailing <code class="inline">\<span class="w">n</span></code>
 character and thus 
trims in a portable way.</p>
<p>When dealing with binary files (or text files in binary mode) be sure
to explicitly set $/ to the appropriate value for your file format
before using chomp().</p>
<p>Because of the &quot;text&quot; mode translation, DOSish perls have limitations
in using <code class="inline"><a class="l_k" href="functions/seek.html">seek</a></code> and <code class="inline"><a class="l_k" href="functions/tell.html">tell</a></code> on a file accessed in &quot;text&quot; mode.
Stick to <code class="inline"><a class="l_k" href="functions/seek.html">seek</a></code>-ing to locations you got from <code class="inline"><a class="l_k" href="functions/tell.html">tell</a></code> (and no
others), and you are usually free to use <code class="inline"><a class="l_k" href="functions/seek.html">seek</a></code> and <code class="inline"><a class="l_k" href="functions/tell.html">tell</a></code> even
in &quot;text&quot; mode.  Using <code class="inline"><a class="l_k" href="functions/seek.html">seek</a></code> or <code class="inline"><a class="l_k" href="functions/tell.html">tell</a></code> or other file operations
may be non-portable.  If you use <code class="inline"><a class="l_k" href="functions/binmode.html">binmode</a></code> on a file, however, you
can usually <code class="inline"><a class="l_k" href="functions/seek.html">seek</a></code> and <code class="inline"><a class="l_k" href="functions/tell.html">tell</a></code> with arbitrary values in safety.</p>
<p>A common misconception in socket programming is that <code class="inline">\<span class="w">n</span></code>
 eq <code class="inline">\<span class="n">012</span></code>

everywhere.  When using protocols such as common Internet protocols,
<code class="inline">\<span class="n">012</span></code>
 and <code class="inline">\<span class="n">015</span></code>
 are called for specifically, and the values of
the logical <code class="inline">\<span class="w">n</span></code>
 and <code class="inline">\<span class="w">r</span></code>
 (carriage return) are not reliable.</p>
<pre class="verbatim"><ol><li>    <a class="l_k" href="functions/print.html">print</a> <span class="i">SOCKET</span> <span class="q">&quot;Hi there, client!\r\n&quot;</span><span class="sc">;</span>      <span class="c"># WRONG</span></li><li>    <a class="l_k" href="functions/print.html">print</a> <span class="i">SOCKET</span> <span class="q">&quot;Hi there, client!\015\012&quot;</span><span class="sc">;</span>  <span class="c"># RIGHT</span></li></ol></pre><p>However, using <code class="inline">\<span class="n">015</span>\<span class="n">012</span></code>
 (or <code class="inline">\<span class="w">cM</span>\<span class="w">cJ</span></code>
, or <code class="inline">\<span class="w">x0D</span>\<span class="w">x0A</span></code>
) 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.</p>
<pre class="verbatim"><ol><li>    <a class="l_k" href="functions/use.html">use</a> <span class="w">Socket</span> <span class="q">qw(:DEFAULT :crlf)</span><span class="sc">;</span></li><li>    <a class="l_k" href="functions/print.html">print</a> <span class="i">SOCKET</span> <span class="q">&quot;Hi there, client!$CRLF&quot;</span>      <span class="c"># RIGHT</span></li></ol></pre><p>When reading from a socket, remember that the default input record
separator <code class="inline"><span class="i">$/</span></code>
 is <code class="inline">\<span class="w">n</span></code>
, but robust socket code will recognize as
either <code class="inline">\<span class="n">012</span></code>
 or <code class="inline">\<span class="n">015</span>\<span class="n">012</span></code>
 as end of line:</p>
<pre class="verbatim"><ol><li>    while <span class="s">(</span><span class="q">&lt;SOCKET&gt;</span><span class="s">)</span> <span class="s">{</span></li><li>        <span class="c"># ...</span></li><li>    <span class="s">}</span></li></ol></pre><p>Because both CRLF and LF end in LF, the input record separator can
be set to LF and any CR stripped later.  Better to write:</p>
<pre class="verbatim"><ol><li>    <a class="l_k" href="functions/use.html">use</a> <span class="w">Socket</span> <span class="q">qw(:DEFAULT :crlf)</span><span class="sc">;</span></li><li>    <a class="l_k" href="functions/local.html">local</a><span class="s">(</span><span class="i">$/</span><span class="s">)</span> = <span class="w">LF</span><span class="sc">;</span>      <span class="c"># not needed if $/ is already \012</span></li><li></li><li>    while <span class="s">(</span><span class="q">&lt;SOCKET&gt;</span><span class="s">)</span> <span class="s">{</span></li><li>        <span class="q">s/$CR?$LF/\n/</span><span class="sc">;</span>   <span class="c"># not sure if socket uses LF or CRLF, OK</span></li><li>    <span class="c">#   s/\015?\012/\n/; # same thing</span></li><li>    <span class="s">}</span></li></ol></pre><p>This example is preferred over the previous one--even for Unix
platforms--because now any <code class="inline">\<span class="n">015</span></code>
&#39;s (<code class="inline">\<span class="w">cM</span></code>
&#39;s) are stripped out
(and there was much rejoicing).</p>
<p>Similarly, functions that return text data--such as a function that
fetches a web page--should sometimes translate newlines before
returning the data, if they&#39;ve not yet been translated to the local
newline representation.  A single line of code will often suffice:</p>
<pre class="verbatim"><ol><li>    <span class="i">$data</span> =~ <span class="q">s/\015?\012/\n/g</span><span class="sc">;</span></li><li>    <a class="l_k" href="functions/return.html">return</a> <span class="i">$data</span><span class="sc">;</span></li></ol></pre><p>Some of this may be confusing.  Here&#39;s a handy reference to the ASCII CR
and LF characters.  You can print it out and stick it in your wallet.</p>
<pre class="verbatim"><ol><li>    LF  eq  \012  eq  \x0A  eq  \cJ  eq  chr(10)  eq  ASCII 10</li><li>    CR  eq  \015  eq  \x0D  eq  \cM  eq  chr(13)  eq  ASCII 13</li><li></li><li>             | Unix | DOS  | Mac  |</li><li>        ---------------------------</li><li>        \n   |  LF  |  LF  |  CR  |</li><li>        \r   |  CR  |  CR  |  LF  |</li><li>        \n * |  LF  | CRLF |  CR  |</li><li>        \r * |  CR  |  CR  |  LF  |</li><li>        ---------------------------</li><li>        * text-mode STDIO</li></ol></pre><p>The Unix column assumes that you are not accessing a serial line
(like a tty) in canonical mode.  If you are, then CR on input becomes
&quot;\n&quot;, and &quot;\n&quot; on output becomes CRLF.</p>
<p>These are just the most common definitions of <code class="inline">\<span class="w">n</span></code>
 and <code class="inline">\<span class="w">r</span></code>
 in 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 &quot;Unix&quot; but the code numbers change:</p>
<pre class="verbatim"><ol><li>    LF  eq  \025  eq  \x15  eq  \cU  eq  chr(21)  eq  CP-1047 21</li><li>    LF  eq  \045  eq  \x25  eq           chr(37)  eq  CP-0037 37</li><li>    CR  eq  \015  eq  \x0D  eq  \cM  eq  chr(13)  eq  CP-1047 13</li><li>    CR  eq  \015  eq  \x0D  eq  \cM  eq  chr(13)  eq  CP-0037 13</li><li></li><li>             | z/OS | OS/400 |</li><li>        ----------------------</li><li>        \n   |  LF  |  LF    |</li><li>        \r   |  CR  |  CR    |</li><li>        \n * |  LF  |  LF    |</li><li>        \r * |  CR  |  CR    |</li><li>        ----------------------</li><li>        * text-mode STDIO</li></ol></pre><a name="Numbers-endianness-and-Width"></a><h2>Numbers endianness and Width</h2>
<p>Different CPUs store integers and floating point numbers in different
orders (called <i>endianness</i>) and widths (32-bit and 64-bit being the
most common today).  This affects your programs when they attempt to transfer
numbers in binary format from one CPU architecture to another,
usually either &quot;live&quot; via network connection, or by storing the
numbers to secondary storage such as a disk file or tape.</p>
<p>Conflicting storage orders make utter mess out of the numbers.  If a
little-endian host (Intel, VAX) stores 0x12345678 (305419896 in
decimal), a big-endian host (Motorola, Sparc, PA) reads it as
0x78563412 (2018915346 in decimal).  Alpha and MIPS can be either:
Digital/Compaq used/uses them in little-endian mode; SGI/Cray uses
them in big-endian mode.  To avoid this problem in network (socket)
connections use the <code class="inline"><a class="l_k" href="functions/pack.html">pack</a></code> and <code class="inline"><a class="l_k" href="functions/unpack.html">unpack</a></code> formats <code class="inline"><span class="w">n</span></code>
 and <code class="inline"><span class="w">N</span></code>
, the
&quot;network&quot; orders.  These are guaranteed to be portable.</p>
<p>As of perl 5.9.2, you can also use the <code class="inline">&gt;</code> and <code class="inline">&lt;</code>
 modifiers
to force big- or little-endian byte-order.  This is useful if you want
to store signed integers or 64-bit integers, for example.</p>
<p>You can explore the endianness of your platform by unpacking a
data structure packed in native format such as:</p>
<pre class="verbatim"><ol><li>    <a class="l_k" href="functions/print.html">print</a> <a class="l_k" href="functions/unpack.html">unpack</a><span class="s">(</span><span class="q">&quot;h*&quot;</span><span class="cm">,</span> <a class="l_k" href="functions/pack.html">pack</a><span class="s">(</span><span class="q">&quot;s2&quot;</span><span class="cm">,</span> <span class="n">1</span><span class="cm">,</span> <span class="n">2</span><span class="s">)</span><span class="s">)</span><span class="cm">,</span> <span class="q">&quot;\n&quot;</span><span class="sc">;</span></li><li>    <span class="c"># &#39;10002000&#39; on e.g. Intel x86 or Alpha 21064 in little-endian mode</span></li><li>    <span class="c"># &#39;00100020&#39; on e.g. Motorola 68040</span></li></ol></pre><p>If you need to distinguish between endian architectures you could use
either of the variables set like so:</p>
<pre class="verbatim"><ol><li>    <span class="i">$is_big_endian</span>   = <a class="l_k" href="functions/unpack.html">unpack</a><span class="s">(</span><span class="q">&quot;h*&quot;</span><span class="cm">,</span> <a class="l_k" href="functions/pack.html">pack</a><span class="s">(</span><span class="q">&quot;s&quot;</span><span class="cm">,</span> <span class="n">1</span><span class="s">)</span><span class="s">)</span> =~ <span class="q">/01/</span><span class="sc">;</span></li><li>    <span class="i">$is_little_endian</span> = <a class="l_k" href="functions/unpack.html">unpack</a><span class="s">(</span><span class="q">&quot;h*&quot;</span><span class="cm">,</span> <a class="l_k" href="functions/pack.html">pack</a><span class="s">(</span><span class="q">&quot;s&quot;</span><span class="cm">,</span> <span class="n">1</span><span class="s">)</span><span class="s">)</span> =~ <span class="q">/^1/</span><span class="sc">;</span></li></ol></pre><p>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 binary numbers.</p>
<p>One can circumnavigate both these problems in two ways.  Either
transfer and store numbers always in text format, instead of raw
binary, or else consider using modules like Data::Dumper (included in
the standard distribution as of Perl 5.005) and Storable (included as
of perl 5.8).  Keeping all data as text significantly simplifies matters.</p>
<p>The v-strings are portable only up to v2147483647 (0x7FFFFFFF), that&#39;s
how far EBCDIC, or more precisely UTF-EBCDIC will go.</p>
<a name="Files-and-Filesystems"></a><h2>Files and Filesystems</h2>
<p>Most platforms these days structure files in a hierarchical fashion.
So, it is reasonably safe to assume that all platforms support the
notion of a &quot;path&quot; to uniquely identify a file on the system.  How
that path is really written, though, differs considerably.</p>
<p>Although similar, file path specifications differ between Unix,
Windows, 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.</p>
<p>DOS, OS/2, VMS, VOS, and Windows can work similarly to Unix with <code class="inline">/</code>
as path separator, or in their own idiosyncratic ways (such as having
several root directories and various &quot;unrooted&quot; device files such NIL:
and LPT:).</p>
<p>Mac OS uses <code class="inline"><span class="j">:</span></code>
 as a path separator instead of <code class="inline">/</code>.</p>
<p>The filesystem may support neither hard links (<code class="inline"><a class="l_k" href="functions/link.html">link</a></code>) nor
symbolic links (<code class="inline"><a class="l_k" href="functions/symlink.html">symlink</a></code>, <code class="inline"><a class="l_k" href="functions/readlink.html">readlink</a></code>, <code class="inline"><a class="l_k" href="functions/lstat.html">lstat</a></code>).</p>
<p>The filesystem may support neither access timestamp nor change
timestamp (meaning that about the only portable timestamp is the
modification timestamp), or one second granularity of any timestamps
(e.g. the FAT filesystem limits the time granularity to two seconds).</p>
<p>The &quot;inode change timestamp&quot; (the <code class="inline">-C</code>
 filetest) may really be the
&quot;creation timestamp&quot; (which it is not in UNIX).</p>
<p>VOS perl can emulate Unix filenames with <code class="inline">/</code> as path separator.  The
native pathname characters greater-than, less-than, number-sign, and
percent-sign are always accepted.</p>
<p>RISC OS perl can emulate Unix filenames with <code class="inline">/</code> as path
separator, or go native and use <code class="inline">.</code> for path separator and <code class="inline"><span class="j">:</span></code>
 to
signal filesystems and disk names.</p>
<p>Don&#39;t assume UNIX filesystem access semantics: that read, write,
and execute are all the permissions there are, and even if 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() work, but sometimes
there simply is no good mapping.</p>
<p>If all this is intimidating, have no (well, maybe only a little)
fear.  There are modules that can help.  The File::Spec modules
provide methods to do the Right Thing on whatever platform happens
to be running the program.</p>
<pre class="verbatim"><ol><li>    <a class="l_k" href="functions/use.html">use</a> <span class="w">File::Spec::Functions</span><span class="sc">;</span></li><li>    <a class="l_k" href="functions/chdir.html">chdir</a><span class="s">(</span><span class="i">updir</span><span class="s">(</span><span class="s">)</span><span class="s">)</span><span class="sc">;</span>        <span class="c"># go up one directory</span></li><li>    <span class="i">$file</span> = <span class="i">catfile</span><span class="s">(</span><span class="i">curdir</span><span class="s">(</span><span class="s">)</span><span class="cm">,</span> <span class="q">&#39;temp&#39;</span><span class="cm">,</span> <span class="q">&#39;file.txt&#39;</span><span class="s">)</span><span class="sc">;</span></li><li>    <span class="c"># on Unix and Win32, &#39;./temp/file.txt&#39;</span></li><li>    <span class="c"># on Mac OS, &#39;:temp:file.txt&#39;</span></li><li>    <span class="c"># on VMS, &#39;[.temp]file.txt&#39;</span></li></ol></pre><p>File::Spec is available in the standard distribution as of version
5.004_05.  File::Spec::Functions is only in File::Spec 0.7 and later,
and some versions of perl come with version 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).</p>
<p>In general, production code should not have file paths hardcoded.
Making them user-supplied or read from a configuration file is
better, keeping in mind that file path syntax varies on different
machines.</p>
<p>This is especially noticeable in scripts like Makefiles and test suites,
which often assume <code class="inline">/</code> as a path separator for subdirectories.</p>
<p>Also of use is File::Basename from the standard distribution, which
splits a pathname into pieces (base filename, full path to directory,
and file suffix).</p>
<p>Even when on a single platform (if you can call Unix a single platform),
remember not to count on the existence or the contents of particular
system-specific files or directories, like <i>/etc/passwd</i>,
<i>/etc/sendmail.conf</i>, <i>/etc/resolv.conf</i>, or even <i>/tmp/</i>.  For
example, <i>/etc/passwd</i> may exist but not contain the encrypted
passwords, because the system is using some form of enhanced security.
Or it may not contain all the accounts, because the system is using NIS. 
If code does need to rely on such a file, include a description of the
file and its format in the code&#39;s documentation, then make it easy for
the user to override the default location of the file.</p>
<p>Don&#39;t assume a text file will end with a newline.  They should,
but people forget.</p>
<p>Do not have two files or directories of the same name with different
case, like <i>test.pl</i> and <i>Test.pl</i>, as many platforms have
case-insensitive (or at least case-forgiving) filenames.  Also, try
not to have non-word characters (except for <code class="inline">.</code>) in the names, and
keep them to the 8.3 convention, for maximum portability, onerous a
burden though this may appear.</p>
<p>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 characters.</p>
<p>Whitespace in 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.</p>
<p>Many systems (DOS, VMS ODS-2) cannot have more than one <code class="inline">.</code> in their
filenames.</p>
<p>Don&#39;t assume <code class="inline">&gt;</code> won&#39;t be the first character of a filename.
Always use <code class="inline">&lt;</code>
 explicitly to open a file for reading, or even
better, use the three-arg version of open, unless you want the user to
be able to specify a pipe open.</p>
<pre class="verbatim"><ol><li>    <a class="l_k" href="functions/open.html">open</a><span class="s">(</span><span class="w">FILE</span><span class="cm">,</span> <span class="q">&#39;&lt;&#39;</span><span class="cm">,</span> <span class="i">$existing_file</span><span class="s">)</span> or <a class="l_k" href="functions/die.html">die</a> <span class="i">$!</span><span class="sc">;</span></li></ol></pre><p>If filenames might use strange characters, it is safest to open it
with <code class="inline"><a class="l_k" href="functions/sysopen.html">sysopen</a></code> instead of <code class="inline"><a class="l_k" href="functions/open.html">open</a></code>.  <code class="inline"><a class="l_k" href="functions/open.html">open</a></code> is magic and can
translate characters like <code class="inline">&gt;</code>, <code class="inline">&lt;</code>
, and <code class="inline">|</code>, which may
be the wrong thing to do.  (Sometimes, though, it&#39;s the right thing.)
Three-arg open can also help protect against this translation in cases
where it is undesirable.</p>
<p>Don&#39;t use <code class="inline"><span class="j">:</span></code>
 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 <code class="inline"><span class="i">@</span></code>
, <code class="inline"><span class="sc">;</span></code>
 and
<code class="inline">|</code>.</p>
<p>Don&#39;t assume that in pathnames you can collapse two leading slashes
<code class="inline"><span class="q">//</span></code>
 into one: some networking and clustering filesystems have special
semantics for that.  Let the operating system to sort it out.</p>
<p>The <i>portable filename characters</i> as defined by ANSI C are</p>
<pre class="verbatim"><ol><li> 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</li><li> 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</li><li> 0 1 2 3 4 5 6 7 8 9</li><li> . _ -</li></ol></pre><p>and the &quot;-&quot; shouldn&#39;t be the first character.  If you want to be
hypercorrect, stay case-insensitive and within the 8.3 naming
convention (all the files and directories have to be unique within one
directory if their names are lowercased and truncated to eight
characters before the <code class="inline">.</code>, if any, and to three characters after the
<code class="inline">.</code>, if any).  (And do not use <code class="inline">.</code>s in directory names.)</p>
<a name="System-Interaction"></a><h2>System Interaction</h2>
<p>Not all platforms provide a command line.  These are usually platforms
that rely primarily on a Graphical User Interface (GUI) for user
interaction.  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&#39;t stay up late worrying about it.</p>
<p>Some platforms can&#39;t delete or rename files held open by the system,
this limitation may also apply to changing filesystem metainformation
like file permissions or owners.  Remember to <code class="inline"><a class="l_k" href="functions/close.html">close</a></code> files when you
are done with them.  Don&#39;t <code class="inline"><a class="l_k" href="functions/unlink.html">unlink</a></code> or <code class="inline"><a class="l_k" href="functions/rename.html">rename</a></code> an open file.  Don&#39;t
<code class="inline"><a class="l_k" href="functions/tie.html">tie</a></code> or <code class="inline"><a class="l_k" href="functions/open.html">open</a></code> a file already tied or opened; <code class="inline"><a class="l_k" href="functions/untie.html">untie</a></code> or <code class="inline"><a class="l_k" href="functions/close.html">close</a></code>
it first.</p>
<p>Don&#39;t open the same file more than once at a time for writing, as some
operating systems put mandatory locks on such files.</p>
<p>Don&#39;t assume that write/modify permission on a directory gives the
right to add or delete files/directories in that directory.  That is
filesystem specific: in some filesystems you need write/modify
permission also (or even just) in the file/directory itself.  In some
filesystems (AFS, DFS) the permission to add/delete directory entries
is a completely separate permission.</p>
<p>Don&#39;t assume that a single <code class="inline"><a class="l_k" href="functions/unlink.html">unlink</a></code> completely gets rid of the file:
some filesystems (most notably the ones in VMS) have versioned
filesystems, and unlink() removes only the most recent one (it doesn&#39;t
remove all the versions because by default the native tools on those
platforms remove just the most recent version, too).  The portable
idiom to remove all the versions of a file is</p>
<pre class="verbatim"><ol><li>    <span class="n">1</span> while <a class="l_k" href="functions/unlink.html">unlink</a> <span class="q">&quot;file&quot;</span><span class="sc">;</span></li></ol></pre><p>This will terminate if the file is undeleteable for some reason
(protected, not there, and so on).</p>
<p>Don&#39;t count on a specific environment variable existing in <code class="inline"><span class="i">%ENV</span></code>
.
Don&#39;t count on <code class="inline"><span class="i">%ENV</span></code>
 entries being case-sensitive, or even
case-preserving.  Don&#39;t try to clear %ENV by saying <code class="inline"><span class="i">%ENV</span> = <span class="s">(</span><span class="s">)</span><span class="sc">;</span></code>
, or,
if you really have to, make it conditional on <code class="inline"><span class="i">$^O</span> ne <span class="q">&#39;VMS&#39;</span></code>
 since in
VMS the <code class="inline"><span class="i">%ENV</span></code>
 table is much more than a per-process key-value string
table.</p>
<p>On VMS, some entries in the %ENV hash are dynamically created when
their key is used on a read if they did not previously exist.  The
values for <code class="inline"><span class="i">$ENV</span>{<span class="w">HOME</span>}</code>
, <code class="inline"><span class="i">$ENV</span>{<span class="w">TERM</span>}</code>
, <code class="inline"><span class="i">$ENV</span>{<span class="w">HOME</span>}</code>
, and <code class="inline"><span class="i">$ENV</span>{<span class="w">USER</span>}</code>
,
are known to be dynamically generated.  The specific names that are
dynamically generated may vary with the version of the C library on VMS,
and more may exist than is documented.</p>
<p>On VMS by default, changes to the %ENV hash are persistent after the process
exits.  This can cause unintended issues.</p>
<p>Don&#39;t count on signals or <code class="inline"><span class="i">%SIG</span></code>
 for anything.</p>
<p>Don&#39;t count on filename globbing.  Use <code class="inline"><a class="l_k" href="functions/opendir.html">opendir</a></code>, <code class="inline"><a class="l_k" href="functions/readdir.html">readdir</a></code>, and
<code class="inline"><a class="l_k" href="functions/closedir.html">closedir</a></code> instead.</p>
<p>Don&#39;t count on per-program environment variables, or per-program current
directories.</p>
<p>Don&#39;t count on specific values of <code class="inline"><span class="i">$!</span></code>
, neither numeric nor
especially the strings values-- users may switch their locales causing
error messages to be translated into their languages.  If you can
trust a POSIXish environment, you can portably use the symbols defined
by the Errno module, like ENOENT.  And don&#39;t trust on the values of <code class="inline"><span class="i">$!</span></code>

at all except immediately after a failed system call.</p>
<a name="Command-names-versus-file-pathnames"></a><h2>Command names versus file pathnames</h2>
<p>Don&#39;t assume that the name used to invoke a command or program with
<code class="inline"><a class="l_k" href="functions/system.html">system</a></code> or <code class="inline"><a class="l_k" href="functions/exec.html">exec</a></code> can also be used to test for the existence of the
file that holds the executable code for that command or program.
First, many systems have &quot;internal&quot; commands that are built-in to the
shell or OS and while these commands can be invoked, there is no
corresponding file.  Second, some operating systems (e.g., Cygwin,
DJGPP, OS/2, and VOS) have required suffixes for executable files;
these suffixes are generally permitted on the command name but are not
required.  Thus, a command like &quot;perl&quot; might exist in a file named
&quot;perl&quot;, &quot;perl.exe&quot;, or &quot;perl.pm&quot;, depending on the operating system.
The variable &quot;_exe&quot; in the Config module holds the executable suffix,
if 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 number in the VMS
file name.</p>
<p>To convert $^X to a file pathname, taking account of the requirements
of the various operating system possibilities, say:</p>
<pre class="verbatim"><ol><li>  <a class="l_k" href="functions/use.html">use</a> <span class="w">Config</span><span class="sc">;</span></li><li>  <span class="i">$thisperl</span> = <span class="i">$^X</span><span class="sc">;</span></li><li>  if <span class="s">(</span><span class="i">$^O</span> ne <span class="q">&#39;VMS&#39;</span><span class="s">)</span></li><li>     <span class="s">{</span><span class="i">$thisperl</span> .= <span class="i">$Config</span>{<span class="w">_exe</span>} unless <span class="i">$thisperl</span> =~ <span class="q">m/$Config{_exe}$/i</span><span class="sc">;</span><span class="s">}</span></li></ol></pre><p>To convert $Config{perlpath} to a file pathname, say:</p>
<pre class="verbatim"><ol><li>  <a class="l_k" href="functions/use.html">use</a> <span class="w">Config</span><span class="sc">;</span></li><li>  <span class="i">$thisperl</span> = <span class="i">$Config</span>{<span class="w">perlpath</span>}<span class="sc">;</span></li><li>  if <span class="s">(</span><span class="i">$^O</span> ne <span class="q">&#39;VMS&#39;</span><span class="s">)</span></li><li>     <span class="s">{</span><span class="i">$thisperl</span> .= <span class="i">$Config</span>{<span class="w">_exe</span>} unless <span class="i">$thisperl</span> =~ <span class="q">m/$Config{_exe}$/i</span><span class="sc">;</span><span class="s">}</span></li></ol></pre><a name="Networking"></a><h2>Networking</h2>
<p>Don&#39;t assume that you can reach the public Internet.</p>
<p>Don&#39;t assume that there is only one way to get through firewalls
to the public Internet.</p>
<p>Don&#39;t assume that you can reach outside world through any other port
than 80, or some web proxy.  ftp is blocked by many firewalls.</p>
<p>Don&#39;t assume that you can send email by connecting to the local SMTP port.</p>
<p>Don&#39;t assume that you can reach yourself or any node by the name
&#39;localhost&#39;.  The same goes for &#39;127.0.0.1&#39;.  You will have to try both.</p>
<p>Don&#39;t assume that the host has only one network card, or that it
can&#39;t bind to many virtual IP addresses.</p>
<p>Don&#39;t assume a particular network device name.</p>
<p>Don&#39;t assume a particular set of ioctl()s will work.</p>
<p>Don&#39;t assume that you can ping hosts and get replies.</p>
<p>Don&#39;t assume that any particular port (service) will respond.</p>
<p>Don&#39;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.</p>
<p>All the above &quot;don&#39;t&quot;:s may look daunting, and they are -- but the key
is to degrade gracefully if one cannot reach the particular network
service one wants.  Croaking or hanging do not look very professional.</p>
<a name="Interprocess-Communication-(IPC)"></a><h2>Interprocess Communication (IPC)</h2>
<p>In general, don&#39;t directly access the system in code meant to be
portable.  That means, no <code class="inline"><a class="l_k" href="functions/system.html">system</a></code>, <code class="inline"><a class="l_k" href="functions/exec.html">exec</a></code>, <code class="inline"><a class="l_k" href="functions/fork.html">fork</a></code>, <code class="inline"><a class="l_k" href="functions/pipe.html">pipe</a></code>,
<code class="inline"><span class="q">``</span></code>
, <code class="inline"><a class="l_k" href="functions/qx.html">qx//</a></code>, <code class="inline"><a class="l_k" href="functions/open.html">open</a></code> with a <code class="inline">|</code>, nor any of the other things
that makes being a perl hacker worth being.</p>
<p>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 differently on different
platforms, may not be available in the same location, might accept
different arguments, can behave differently, and often present their
results in a platform-dependent way.  Thus, you should seldom depend
on them to produce consistent results. (Then again, if you&#39;re calling 
<i>netstat -a</i>, you probably don&#39;t expect it to run on both Unix and CP/M.)</p>
<p>One especially common bit of Perl code is opening a pipe to <b>sendmail</b>:</p>
<pre class="verbatim"><ol><li>    <a class="l_k" href="functions/open.html">open</a><span class="s">(</span><span class="w">MAIL</span><span class="cm">,</span> <span class="q">&#39;|/usr/lib/sendmail -t&#39;</span><span class="s">)</span> </li><li>	or <a class="l_k" href="functions/die.html">die</a> <span class="q">&quot;cannot fork sendmail: $!&quot;</span><span class="sc">;</span></li></ol></pre><p>This is fine for systems programming when sendmail 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 installed.  If a portable
solution is needed, see the various distributions on CPAN that deal
with it.  Mail::Mailer and Mail::Send in the MailTools distribution are
commonly used, and provide several mailing methods, including mail,
sendmail, and direct SMTP (via Net::SMTP) if a mail transfer agent is
not available.  Mail::Sendmail is a standalone module that provides
simple, platform-independent mailing.</p>
<p>The Unix System V IPC (<code class="inline"><span class="w">msg</span>*<span class="s">(</span><span class="s">)</span><span class="cm">,</span> <span class="w">sem</span>*<span class="s">(</span><span class="s">)</span><span class="cm">,</span> <span class="w">shm</span>*<span class="s">(</span><span class="s">)</span></code>
) is not available
even on all Unix platforms.</p>
<p>Do not use either the bare result of <code class="inline"><a class="l_k" href="functions/pack.html">pack</a><span class="s">(</span><span class="q">&quot;N&quot;</span><span class="cm">,</span> <span class="n">10</span><span class="cm">,</span> <span class="n">20</span><span class="cm">,</span> <span class="n">30</span><span class="cm">,</span> <span class="n">40</span><span class="s">)</span></code>
 or
bare v-strings (such as <code class="inline"><span class="v">v10.20.30.40</span></code>
) to represent IPv4 addresses:
both forms just pack the four bytes into network order.  That this
would be equal to the C language <code class="inline"><span class="w">in_addr</span></code>
 struct (which is what the
socket code internally uses) is not guaranteed.  To be portable use
the routines of the Socket extension, such as <code class="inline"><span class="i">inet_aton</span><span class="s">(</span><span class="s">)</span></code>
,
<code class="inline"><span class="i">inet_ntoa</span><span class="s">(</span><span class="s">)</span></code>
, and <code class="inline"><span class="i">sockaddr_in</span><span class="s">(</span><span class="s">)</span></code>
.</p>
<p>The rule of thumb for portable code is: Do it all in portable Perl, or
use a module (that may internally implement it with platform-specific
code, but expose a common interface).</p>
<a name="External-Subroutines-(XS)"></a><h2>External Subroutines (XS)</h2>
<p>XS code can usually be made to work with any platform, but dependent
libraries, header files, etc., might not be readily available or
portable, 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.</p>
<p>A different type of portability issue arises when writing XS code:
availability of a C compiler on the end-user&#39;s system.  C brings
with it its own portability issues, and writing XS code will expose
you to some of those.  Writing purely in Perl is an easier way to
achieve portability.</p>
<a name="Standard-Modules"></a><h2>Standard Modules</h2>
<p>In general, the standard modules work across platforms.  Notable
exceptions are the CPAN module (which currently makes connections to external
programs that may not be available), platform-specific modules (like
ExtUtils::MM_VMS), and DBM modules.</p>
<p>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 MacPerl, where only NBDM_File and DB_File are
available.</p>
<p>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 (e.g., not exceeding 1K for each record), so that it will
work with any DBM module.  See <a href="AnyDBM_File.html">AnyDBM_File</a> for more details.</p>
<a name="Time-and-Date"></a><h2>Time and Date</h2>
<p>The system&#39;s notion of time of day and calendar date is controlled in
widely different ways.  Don&#39;t assume the timezone is stored in <code class="inline"><span class="i">$ENV</span>{<span class="w">TZ</span>}</code>
,
and even if it is, don&#39;t assume that you can control the timezone through
that variable.  Don&#39;t assume anything about the three-letter timezone
abbreviations (for example that MST would be the Mountain Standard Time,
it&#39;s been known to stand for Moscow Standard Time).  If you need to
use timezones, express them in some unambiguous format like the
exact number of minutes offset from UTC, or the POSIX timezone
format.</p>
<p>Don&#39;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 an unambiguous representation.  The ISO 8601 standard
defines YYYY-MM-DD as the date format, or YYYY-MM-DDTHH-MM-SS
(that&#39;s a literal &quot;T&quot; separating the date from the time).
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 &quot;1987-12-18&quot;) can be easily converted
into an OS-specific value using a module like Date::Parse.
An array of values, such as those returned by <code class="inline"><a class="l_k" href="functions/localtime.html">localtime</a></code>, can be
converted to an OS-specific representation using Time::Local.</p>
<p>When calculating specific times, such as for tests in time or date modules,
it may be appropriate to calculate an offset for the epoch.</p>
<pre class="verbatim"><ol><li>    <a class="l_k" href="functions/require.html">require</a> <span class="w">Time::Local</span><span class="sc">;</span></li><li>    <span class="i">$offset</span> = <span class="i">Time::Local::timegm</span><span class="s">(</span><span class="n">0</span><span class="cm">,</span> <span class="n">0</span><span class="cm">,</span> <span class="n">0</span><span class="cm">,</span> <span class="n">1</span><span class="cm">,</span> <span class="n">0</span><span class="cm">,</span> <span class="n">70</span><span class="s">)</span><span class="sc">;</span></li></ol></pre><p>The value for <code class="inline"><span class="i">$offset</span></code>
 in Unix will be <code class="inline"><span class="n">0</span></code>
, but in Mac OS will be
some large number.  <code class="inline"><span class="i">$offset</span></code>
 can then be added to a Unix time value
to get what should be the proper value on any system.</p>
<p>On Windows (at least), you shouldn&#39;t pass a negative value to <code class="inline"><a class="l_k" href="functions/gmtime.html">gmtime</a></code> or
<code class="inline"><a class="l_k" href="functions/localtime.html">localtime</a></code>.</p>
<a name="Character-sets-and-character-encoding"></a><h2>Character sets and character encoding</h2>
<p>Assume very little about character sets.</p>
<p>Assume nothing about numerical values (<code class="inline"><a class="l_k" href="functions/ord.html">ord</a></code>, <code class="inline"><a class="l_k" href="functions/chr.html">chr</a></code>) of characters.
Do not use explicit code point ranges (like \xHH-\xHH); use for
example symbolic character classes like <code class="inline">[:print:]</code>.</p>
<p>Do not assume that the alphabetic characters are encoded contiguously
(in the numeric sense).  There may be gaps.</p>
<p>Do not assume anything about the ordering of the characters.
The lowercase letters may come before or after the uppercase letters;
the lowercase and uppercase may be interlaced so that both &quot;a&quot; and &quot;A&quot;
come before &quot;b&quot;; the accented and other international characters may
be interlaced so that  comes before &quot;b&quot;.</p>
<a name="Internationalisation"></a><h2>Internationalisation</h2>
<p>If you may assume POSIX (a rather large assumption), you may read
more about the POSIX locale system from <a href="perllocale.html">perllocale</a>.  The locale
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, and date
and time formatting--amongst other things.</p>
<p>If you really want to be international, you should consider Unicode.
See <a href="perluniintro.html">perluniintro</a> and <a href="perlunicode.html">perlunicode</a> for more information.</p>
<p>If you want to use non-ASCII bytes (outside the bytes 0x00..0x7f) in
the &quot;source code&quot; 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, in which case random native bytes might be
illegal (&quot;Malformed UTF-8 ...&quot;)  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 <code class="inline"><span class="w">bytes</span></code>

pragma.  If the bytes are in a string (regular expression being a
curious string), you can often also use the <code class="inline">\<span class="w">xHH</span></code>
 notation instead
of embedding the bytes as-is.  (If you want to write your code in UTF-8,
you can use the <code class="inline"><span class="w">utf8</span></code>
.) The <code class="inline"><span class="w">bytes</span></code>
 and <code class="inline"><span class="w">utf8</span></code>
 pragmata are
available since Perl 5.6.0.</p>
<a name="System-Resources"></a><h2>System Resources</h2>
<p>If your code is destined for systems with severely constrained (or
missing!) virtual memory systems then you want to be <i>especially</i> mindful
of avoiding wasteful constructs such as:</p>
<pre class="verbatim"><ol><li>    <span class="c"># NOTE: this is no longer &quot;bad&quot; in perl5.005</span></li><li>    for <span class="s">(</span><span class="n">0</span>..<span class="n">10000000</span><span class="s">)</span> <span class="s">{</span><span class="s">}</span>                       <span class="c"># bad</span></li><li>    for <span class="s">(</span><a class="l_k" href="functions/my.html">my</a> <span class="i">$x</span> = <span class="n">0</span><span class="sc">;</span> <span class="i">$x</span> &lt;= <span class="n">10000000</span><span class="sc">;</span> ++<span class="i">$x</span><span class="s">)</span> <span class="s">{</span><span class="s">}</span>   <span class="c"># good</span></li><li></li><li>    <span class="i">@lines</span> = <span class="q">&lt;VERY_LARGE_FILE&gt;</span><span class="sc">;</span>                <span class="c"># bad</span></li><li></li><li>    while <span class="s">(</span><span class="q">&lt;FILE&gt;</span><span class="s">)</span> <span class="s">{</span><span class="i">$file</span> .= <span class="i">$_</span><span class="s">}</span>               <span class="c"># sometimes bad</span></li><li>    <span class="i">$file</span> = <a class="l_k" href="functions/join.html">join</a><span class="s">(</span><span class="q">&#39;&#39;</span><span class="cm">,</span> <span class="q">&lt;FILE&gt;</span><span class="s">)</span><span class="sc">;</span>                  <span class="c"># better</span></li></ol></pre><p>The last two constructs may appear unintuitive to most people.  The
first repeatedly grows a string, whereas the second allocates a
large chunk of memory in one go.  On some systems, the second is
more efficient that the first.</p>
<a name="Security"></a><h2>Security</h2>
<p>Most multi-user platforms provide basic levels of security, usually
implemented at the filesystem level.  Some, however, do
not-- unfortunately.  Thus the notion of user id, or &quot;home&quot; directory,
or even the state of being logged-in, may be unrecognizable on many
platforms.  If you write 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 code explicitly for that platform (or
class of platforms).</p>
<p>Don&#39;t assume the UNIX filesystem access semantics: the operating
system or the filesystem may be using some ACL systems, which are
richer languages than the usual rwx.  Even if the rwx exist,
their semantics might be different.</p>
<p>(From security viewpoint testing for permissions before attempting to
do something is silly anyway: if 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 operation.)</p>
<p>Don&#39;t assume the UNIX user and group semantics: especially, don&#39;t
expect the <code class="inline"><span class="i">$&lt;</span></code>
 and <code class="inline"><span class="i">$&gt;</span></code>
 (or the <code class="inline"><span class="i">$(</span></code>
 and <code class="inline"><span class="i">$)</span></code>
) to work
for switching identities (or memberships).</p>
<p>Don&#39;t assume set-uid and set-gid semantics. (And even if you do,
think twice: set-uid and set-gid are a known can of security worms.)</p>
<a name="Style"></a><h2>Style</h2>
<p>For those times when it is necessary to have platform-specific code,
consider keeping the platform-specific code in one place, making porting
to other platforms easier.  Use the Config module and the special
variable <code class="inline"><span class="i">$^O</span></code>
 to differentiate platforms, as described in
<a href="http://search.cpan.org/perldoc/&quot;PLATFORMS&quot;">PLATFORMS</a>.</p>
<p>Be careful in the tests you supply with your module or programs.
Module code may be fully portable, but its tests might not be.  This
often happens when tests spawn off other processes or call external
programs to aid in the testing, or when (as noted above) the tests
assume certain things about the filesystem and paths.  Be careful not
to depend on a specific output style for errors, such as when checking
<code class="inline"><span class="i">$!</span></code>
 after a failed system call.  Using <code class="inline"><span class="i">$!</span></code>
 for anything else than
displaying it as output is doubtful (though see the Errno module for
testing reasonably portably for error value). Some platforms expect
a certain output format, and Perl on those platforms may have been
adjusted accordingly.  Most specifically, don&#39;t anchor a regex when
testing an error value.</p>
<a name="CPAN-Testers"></a><h1>CPAN Testers</h1>
<p>Modules uploaded to CPAN are tested by a variety of volunteers on
different platforms.  These CPAN testers are notified by mail of each
new upload, and reply to the list with PASS, FAIL, NA (not applicable to
this platform), or UNKNOWN (unknown), along with any relevant notations.</p>
<p>The purpose of the testing is twofold: one, to help developers fix any
problems in their code that crop up because of lack of testing on other
platforms; two, to provide users with information about whether
a given module works on a given platform.</p>
<p>Also see:</p>
<ul>
<li>
<p>Mailing list: cpan-testers@perl.org</p>
</li>
<li>
<p>Testing results: <a href="http://testers.cpan.org/">http://testers.cpan.org/</a></p>
</li>
</ul>
<a name="PLATFORMS"></a><h1>PLATFORMS</h1>
<p>As of version 5.002, Perl is built with a <code class="inline"><span class="i">$^O</span></code>
 variable that
indicates the operating system it was built on.  This was implemented
to help speed up code that would otherwise have to <code class="inline"><a class="l_k" href="functions/use.html">use</a> <span class="w">Config</span></code>

and use the value of <code class="inline"><span class="i">$Config</span>{<span class="w">osname</span>}</code>
.  Of course, to get more
detailed information about the system, looking into <code class="inline"><span class="i">%Config</span></code>
 is
certainly recommended.</p>
<p><code class="inline"><span class="i">%Config</span></code>
 cannot always be trusted, however, because it was built
at compile time.  If perl was built in one place, then transferred
elsewhere, some values may be wrong.  The values may even have been
edited after the fact.</p>
<a name="Unix"></a><h2>Unix</h2>
<p>Perl works on a bewildering variety of Unix and Unix-like platforms (see
e.g. most of the files in the <i>hints/</i> directory in the source code kit).
On most of these systems, the value of <code class="inline"><span class="i">$^O</span></code>
 (hence <code class="inline"><span class="i">$Config</span>{<span class="q">&#39;osname&#39;</span>}</code>
,
too) is determined either by lowercasing and stripping punctuation from the
first field of the string returned by typing <code class="inline"><span class="w">uname</span> -<span class="w">a</span></code>
 (or a similar command)
at the shell prompt or by testing the file system for the presence of
uniquely named files such as a kernel or header file.  Here, for example,
are a few of the more popular Unix flavors:</p>
<pre class="verbatim"><ol><li>    uname         $^O        $Config{'archname'}</li><li>    --------------------------------------------</li><li>    AIX           aix        aix</li><li>    BSD/OS        bsdos      i386-bsdos</li><li>    Darwin        darwin     darwin</li><li>    dgux          dgux       AViiON-dgux</li><li>    DYNIX/ptx     dynixptx   i386-dynixptx</li><li>    FreeBSD       freebsd    freebsd-i386    </li><li>    Haiku         haiku      BePC-haiku</li><li>    Linux         linux      arm-linux</li><li>    Linux         linux      i386-linux</li><li>    Linux         linux      i586-linux</li><li>    Linux         linux      ppc-linux</li><li>    HP-UX         hpux       PA-RISC1.1</li><li>    IRIX          irix       irix</li><li>    Mac OS X      darwin     darwin</li><li>    MachTen PPC   machten    powerpc-machten</li><li>    NeXT 3        next       next-fat</li><li>    NeXT 4        next       OPENSTEP-Mach</li><li>    openbsd       openbsd    i386-openbsd</li><li>    OSF1          dec_osf    alpha-dec_osf</li><li>    reliantunix-n svr4       RM400-svr4</li><li>    SCO_SV        sco_sv     i386-sco_sv</li><li>    SINIX-N       svr4       RM400-svr4</li><li>    sn4609        unicos     CRAY_C90-unicos</li><li>    sn6521        unicosmk   t3e-unicosmk</li><li>    sn9617        unicos     CRAY_J90-unicos</li><li>    SunOS         solaris    sun4-solaris</li><li>    SunOS         solaris    i86pc-solaris</li><li>    SunOS4        sunos      sun4-sunos</li></ol></pre><p>Because the value of <code class="inline"><span class="i">$Config</span>{<span class="w">archname</span>}</code>
 may depend on the
hardware architecture, it can vary more than the value of <code class="inline"><span class="i">$^O</span></code>
.</p>
<a name="DOS-and-Derivatives"></a><h2>DOS and Derivatives</h2>
<p>Perl has long been ported to Intel-style microcomputers running under
systems like PC-DOS, MS-DOS, OS/2, and most Windows platforms you can
bring yourself to mention (except for Windows CE, if you count that).
Users familiar with <i>COMMAND.COM</i> or <i>CMD.EXE</i> style shells should
be aware that each of these file specifications may have subtle
differences:</p>
<pre class="verbatim"><ol><li>    <span class="i">$filespec0</span> = <span class="q">&quot;c:/foo/bar/file.txt&quot;</span><span class="sc">;</span></li><li>    <span class="i">$filespec1</span> = <span class="q">&quot;c:\\foo\\bar\\file.txt&quot;</span><span class="sc">;</span></li><li>    <span class="i">$filespec2</span> = <span class="q">&#39;c:\foo\bar\file.txt&#39;</span><span class="sc">;</span></li><li>    <span class="i">$filespec3</span> = <span class="q">&#39;c:\\foo\\bar\\file.txt&#39;</span><span class="sc">;</span></li></ol></pre><p>System calls accept either <code class="inline">/</code> or <code class="inline">\</code>
 as the path separator.
However, many command-line utilities of DOS vintage treat <code class="inline">/</code> as
the option prefix, so may get confused by filenames containing <code class="inline">/</code>.
Aside from calling any external programs, <code class="inline">/</code> will work just fine,
and probably better, as it is more consistent with popular usage,
and avoids the problem of remembering what to backwhack and what
not to.</p>
<p>The DOS FAT filesystem can accommodate only &quot;8.3&quot; style filenames.  Under
the &quot;case-insensitive, but case-preserving&quot; HPFS (OS/2) and NTFS (NT)
filesystems you may have to be careful about case returned with functions
like <code class="inline"><a class="l_k" href="functions/readdir.html">readdir</a></code> or used with functions like <code class="inline"><a class="l_k" href="functions/open.html">open</a></code> or <code class="inline"><a class="l_k" href="functions/opendir.html">opendir</a></code>.</p>
<p>DOS also treats several filenames as special, such as AUX, PRN,
NUL, CON, COM1, LPT1, LPT2, etc.  Unfortunately, sometimes these
filenames won&#39;t even work if you include an explicit directory
prefix.  It is best to avoid such filenames, if you want your code
to be portable to DOS and its derivatives.  It&#39;s hard to know what
these all are, unfortunately.</p>
<p>Users of these operating systems may also wish to make use of
scripts such as <i>pl2bat.bat</i> or <i>pl2cmd</i> to
put wrappers around your scripts.</p>
<p>Newline (<code class="inline">\<span class="w">n</span></code>
) is translated as <code class="inline">\<span class="n">015</span>\<span class="n">012</span></code>
 by STDIO when reading from
and writing to files (see <a href="http://search.cpan.org/perldoc/&quot;Newlines&quot;">Newlines</a>).  <code class="inline"><a class="l_k" href="functions/binmode.html">binmode(FILEHANDLE)</a></code>
will keep <code class="inline">\<span class="w">n</span></code>
 translated as <code class="inline">\<span class="n">012</span></code>
 for that filehandle.  Since it is a
no-op on other systems, <code class="inline"><a class="l_k" href="functions/binmode.html">binmode</a></code> should be used for cross-platform code
that deals with binary data.  That&#39;s assuming you realize in advance
that your data is in binary.  General-purpose programs should
often assume nothing about their data.</p>
<p>The <code class="inline"><span class="i">$^O</span></code>
 variable and the <code class="inline"><span class="i">$Config</span>{<span class="w">archname</span>}</code>
 values for various
DOSish perls are as follows:</p>
<pre class="verbatim"><ol><li>     OS            $^O      $Config{archname}   ID    Version</li><li>     --------------------------------------------------------</li><li>     MS-DOS        dos        ?                 </li><li>     PC-DOS        dos        ?                 </li><li>     OS/2          os2        ?</li><li>     Windows 3.1   ?          ?                 0      3 01</li><li>     Windows 95    MSWin32    MSWin32-x86       1      4 00</li><li>     Windows 98    MSWin32    MSWin32-x86       1      4 10</li><li>     Windows ME    MSWin32    MSWin32-x86       1      ?</li><li>     Windows NT    MSWin32    MSWin32-x86       2      4 xx</li><li>     Windows NT    MSWin32    MSWin32-ALPHA     2      4 xx</li><li>     Windows NT    MSWin32    MSWin32-ppc       2      4 xx</li><li>     Windows 2000  MSWin32    MSWin32-x86       2      5 00</li><li>     Windows XP    MSWin32    MSWin32-x86       2      5 01</li><li>     Windows 2003  MSWin32    MSWin32-x86       2      5 02</li><li>     Windows CE    MSWin32    ?                 3           </li><li>     Cygwin        cygwin     cygwin</li></ol></pre><p>The various MSWin32 Perl&#39;s can distinguish the OS they are running on
via the value of the fifth element of the list returned from 
Win32::GetOSVersion().  For example:</p>
<pre class="verbatim"><ol><li>    if <span class="s">(</span><span class="i">$^O</span> eq <span class="q">&#39;MSWin32&#39;</span><span class="s">)</span> <span class="s">{</span></li><li>        <a class="l_k" href="functions/my.html">my</a> <span class="i">@os_version_info</span> = <span class="i">Win32::GetOSVersion</span><span class="s">(</span><span class="s">)</span><span class="sc">;</span></li><li>        <a class="l_k" href="functions/print.html">print</a> +<span class="s">(</span><span class="q">&#39;3.1&#39;</span><span class="cm">,</span><span class="q">&#39;95&#39;</span><span class="cm">,</span><span class="q">&#39;NT&#39;</span><span class="s">)</span>[<span class="i">$os_version_info</span>[<span class="n">4</span>]]<span class="cm">,</span><span class="q">&quot;\n&quot;</span><span class="sc">;</span></li><li>    <span class="s">}</span></li></ol></pre><p>There are also Win32::IsWinNT() and Win32::IsWin95(), try <code class="inline"><span class="w">perldoc</span> <span class="w">Win32</span></code>
,
and as of libwin32 0.19 (not part of the core Perl distribution)
Win32::GetOSName().  The very portable POSIX::uname() will work too:</p>
<pre class="verbatim"><ol><li>    c:\&gt; perl -MPOSIX -we "print join '|', uname"</li><li>    Windows NT|moonru|5.0|Build 2195 (Service Pack 2)|x86</li></ol></pre><p>Also see:</p>
<ul>
<li>
<p>The djgpp environment for DOS, <a href="http://www.delorie.com/djgpp/">http://www.delorie.com/djgpp/</a>
and <a href="perldos.html">perldos</a>.</p>
</li>
<li>
<p>The EMX environment for DOS, OS/2, etc. emx@iaehv.nl,
<a href="http://www.leo.org/pub/comp/os/os2/leo/gnu/emx+gcc/index.html">http://www.leo.org/pub/comp/os/os2/leo/gnu/emx+gcc/index.html</a> or
<a href="ftp://hobbes.nmsu.edu/pub/os2/dev/emx/">ftp://hobbes.nmsu.edu/pub/os2/dev/emx/</a>  Also <a href="perlos2.html">perlos2</a>.</p>
</li>
<li>
<p>Build instructions for Win32 in <a href="perlwin32.html">perlwin32</a>, or under the Cygnus environment
in <a href="perlcygwin.html">perlcygwin</a>.</p>
</li>
<li>
<p>The <code class="inline"><span class="w">Win32::</span>*</code>
 modules in <a href="http://search.cpan.org/perldoc/Win32">Win32</a>.</p>
</li>
<li>
<p>The ActiveState Pages, <a href="http://www.activestate.com/">http://www.activestate.com/</a></p>
</li>
<li>
<p>The Cygwin environment for Win32; <i>README.cygwin</i> (installed 
as <a href="perlcygwin.html">perlcygwin</a>), <a href="http://www.cygwin.com/">http://www.cygwin.com/</a></p>
</li>
<li>
<p>The U/WIN environment for Win32,
<a href="http://www.research.att.com/sw/tools/uwin/">http://www.research.att.com/sw/tools/uwin/</a></p>
</li>
<li>
<p>Build instructions for OS/2, <a href="perlos2.html">perlos2</a></p>
</li>
</ul>
<a name="Mac-OS"></a><h2>Mac OS</h2>
<p>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 binary
form on CPAN.</p>
<p>Directories are specified as:</p>
<pre class="verbatim"><ol><li>    volume:folder:file              for absolute pathnames</li><li>    volume:folder:                  for absolute pathnames</li><li>    :folder:file                    for relative pathnames</li><li>    :folder:                        for relative pathnames</li><li>    :file                           for relative pathnames</li><li>    file                            for relative pathnames</li></ol></pre><p>Files are stored in the directory in alphabetical order.  Filenames are
limited to 31 characters, and may include any character except for
null and <code class="inline"><span class="j">:</span></code>
, which is reserved as the path separator.</p>
<p>Instead of <code class="inline"><a class="l_k" href="functions/flock.html">flock</a></code>, see <code class="inline"><span class="w">FSpSetFLock</span></code>
 and <code class="inline"><span class="w">FSpRstFLock</span></code>
 in the
Mac::Files module, or <code class="inline"><a class="l_k" href="functions/chmod.html">chmod</a><span class="s">(</span><span class="n">0444</span><span class="cm">,</span> ...<span class="s">)</span></code>
 and <code class="inline"><a class="l_k" href="functions/chmod.html">chmod</a><span class="s">(</span><span class="n">0666</span><span class="cm">,</span> ...<span class="s">)</span></code>
.</p>
<p>In the MacPerl application, you can&#39;t run a program from the command line;
programs that expect <code class="inline"><span class="i">@ARGV</span></code>
 to be populated can be edited with something
like the following, which brings up a dialog box asking for the command
line arguments.</p>
<pre class="verbatim"><ol><li>    if <span class="s">(</span>!<span class="i">@ARGV</span><span class="s">)</span> <span class="s">{</span></li><li>        <span class="i">@ARGV</span> = <a class="l_k" href="functions/split.html">split</a> <span class="q">/\s+/</span><span class="cm">,</span> <span class="i">MacPerl::Ask</span><span class="s">(</span><span class="q">&#39;Arguments?&#39;</span><span class="s">)</span><span class="sc">;</span></li><li>    <span class="s">}</span></li></ol></pre><p>A MacPerl script saved as a &quot;droplet&quot; will populate <code class="inline"><span class="i">@ARGV</span></code>
 with the full
pathnames of the files dropped onto the script.</p>
<p>Mac users can run programs under a type of command line interface
under MPW (Macintosh Programmer&#39;s Workshop, a free development
environment from Apple).  MacPerl was first introduced as an MPW
tool, and MPW can be used like a shell:</p>
<pre class="verbatim"><ol><li>    <span class="w">perl</span> <span class="w">myscript</span>.<span class="w">plx</span> <span class="w">some</span> <span class="w">arguments</span></li></ol></pre><p>ToolServer is another app from Apple that provides access to MPW tools
from MPW and the MacPerl app, which allows MacPerl programs to use
<code class="inline"><a class="l_k" href="functions/system.html">system</a></code>, backticks, and piped <code class="inline"><a class="l_k" href="functions/open.html">open</a></code>.</p>
<p>&quot;Mac OS&quot; is the proper name for the operating system, but the value
in <code class="inline"><span class="i">$^O</span></code>
 is &quot;MacOS&quot;.  To determine architecture, version, or whether
the application or MPW tool version is running, check:</p>
<pre class="verbatim"><ol><li>    <span class="i">$is_app</span>    = <span class="i">$MacPerl::Version</span> =~ <span class="q">/App/</span><span class="sc">;</span></li><li>    <span class="i">$is_tool</span>   = <span class="i">$MacPerl::Version</span> =~ <span class="q">/MPW/</span><span class="sc">;</span></li><li>    <span class="s">(</span><span class="i">$version</span><span class="s">)</span> = <span class="i">$MacPerl::Version</span> =~ <span class="q">/^(\S+)/</span><span class="sc">;</span></li><li>    <span class="i">$is_ppc</span>    = <span class="i">$MacPerl::Architecture</span> eq <span class="q">&#39;MacPPC&#39;</span><span class="sc">;</span></li><li>    <span class="i">$is_68k</span>    = <span class="i">$MacPerl::Architecture</span> eq <span class="q">&#39;Mac68K&#39;</span><span class="sc">;</span></li></ol></pre><p>Mac OS X, based on NeXT&#39;s OpenStep OS, runs MacPerl natively, under the
&quot;Classic&quot; environment.  There is no &quot;Carbon&quot; version of MacPerl to run
under the primary Mac OS X environment.  Mac OS X and its Open Source
version, Darwin, both run Unix perl natively.</p>
<p>Also see:</p>
<ul>
<li>
<p>MacPerl Development, <a href="http://dev.macperl.org/">http://dev.macperl.org/</a> .</p>
</li>
<li>
<p>The MacPerl Pages, <a href="http://www.macperl.com/">http://www.macperl.com/</a> .</p>
</li>
<li>
<p>The MacPerl mailing lists, <a href="http://lists.perl.org/">http://lists.perl.org/</a> .</p>
</li>
<li>
<p>MPW, <a href="ftp://ftp.apple.com/developer/Tool_Chest/Core_Mac_OS_Tools/">ftp://ftp.apple.com/developer/Tool_Chest/Core_Mac_OS_Tools/</a></p>
</li>
</ul>
<a name="VMS"></a><h2>VMS</h2>
<p>Perl on VMS is discussed in <a href="perlvms.html">perlvms</a> in the perl distribution.</p>
<p>The official name of VMS as of this writing is OpenVMS.</p>
<p>Perl on VMS can accept either VMS- or Unix-style file
specifications as in either of the following:</p>
<pre class="verbatim"><ol><li>    $ perl -ne "print if /perl_setup/i" SYS$LOGIN:LOGIN.COM</li><li>    $ perl -ne "print if /perl_setup/i" /sys$login/login.com</li></ol></pre><p>but not a mixture of both as in:</p>
<pre class="verbatim"><ol><li>    $ perl -ne "print if /perl_setup/i" sys$login:/login.com</li><li>    Can't open sys$login:/login.com: file specification syntax error</li></ol></pre><p>Interacting with Perl from the Digital Command Language (DCL) shell
often requires a different set of quotation marks than Unix shells do.
For example:</p>
<pre class="verbatim"><ol><li>    $ perl -e "print ""Hello, world.\n"""</li><li>    Hello, world.</li></ol></pre><p>There are several ways to wrap your perl scripts in DCL <i>.COM</i> files, if
you are so inclined.  For example:</p>
<pre class="verbatim"><ol><li>    $ write sys$output "Hello from DCL!"</li><li>    $ if p1 .eqs. ""</li><li>    $ then perl -x 'f$environment("PROCEDURE")</li><li>    $ else perl -x - 'p1 'p2 'p3 'p4 'p5 'p6 'p7 'p8</li><li>    $ deck/dollars="__END__"</li><li>    #!/usr/bin/perl</li><li></li><li>    print "Hello from Perl!\n";</li><li></li><li>    __END__</li><li>    $ endif</li></ol></pre><p>Do take care with <code class="inline">$ ASSIGN/nolog/user SYS$COMMAND: SYS$INPUT</code> if your
perl-in-DCL script expects to do things like <code class="inline"><span class="i">$read</span> = <span class="q">&lt;STDIN&gt;</span><span class="sc">;</span></code>
.</p>
<p>The VMS operating system has two filesystems, known as ODS-2 and ODS-5.</p>
<p>For ODS-2, filenames are in the format &quot;name.extension;version&quot;.  The
maximum length for filenames is 39 characters, and the maximum length for
extensions is also 39 characters.  Version is a number from 1 to
32767.  Valid characters are <code class="inline"><span class="q">/[A-Z0-9$_-]/</span></code>
.</p>
<p>The ODS-2 filesystem is case-insensitive and does not preserve case.
Perl simulates this by converting all filenames to lowercase internally.</p>
<p>For ODS-5, filenames may have almost any character in them and can include
Unicode characters.  Characters that could be misinterpreted by the DCL
shell or file parsing utilities need to be prefixed with the <code class="inline">^</code>
character, or replaced with hexadecimal characters prefixed with the
<code class="inline">^</code> character.  Such prefixing is only needed with the pathnames are
in VMS format in applications.  Programs that can accept the UNIX format
of pathnames do not need the escape characters.  The maximum length for
filenames is 255 characters.  The ODS-5 file system can handle both
a case preserved and a case sensitive mode.</p>
<p>ODS-5 is only available on the OpenVMS for 64 bit platforms.</p>
<p>Support for the extended file specifications is being done as optional
settings to preserve backward compatibility with Perl scripts that
assume the previous VMS limitations.</p>
<p>In general routines on VMS that get a UNIX format file specification
should return it in a UNIX format, and when they get a VMS format
specification they should return a VMS format unless they are documented
to do a conversion.</p>
<p>For routines that generate return a file specification, VMS allows setting
if the C library which Perl is built on if it will be returned in VMS
format or in UNIX format.</p>
<p>With the ODS-2 file system, there is not much difference in syntax of
filenames without paths for VMS or UNIX.  With the extended character
set available with ODS-5 there can be a significant difference.</p>
<p>Because of this, existing Perl scripts written for VMS were sometimes
treating VMS and UNIX filenames interchangeably.  Without the extended
character set enabled, this behavior will mostly be maintained for
backwards compatibility.</p>
<p>When extended characters are enabled with ODS-5, the handling of
UNIX formatted file specifications is to that of a UNIX system.</p>
<p>VMS file specifications without extensions have a trailing dot.  An
equivalent UNIX file specification should not show the trailing dot.</p>
<p>The result of all of this, is that for VMS, for portable scripts, you
can not depend on Perl to present the filenames in lowercase, to be
case sensitive, and that the filenames could be returned in either
UNIX or VMS format.</p>
<p>And if a routine returns a file specification, unless it is intended to
convert it, it should return it in the same format as it found it.</p>
<p><code class="inline"><a class="l_k" href="functions/readdir.html">readdir</a></code> by default has traditionally returned lowercased filenames.
When the ODS-5 support is enabled, it will return the exact case of the
filename on the disk.</p>
<p>Files without extensions have a trailing period on them, so doing a
<code class="inline"><a class="l_k" href="functions/readdir.html">readdir</a></code> in the default mode with a file named <i>A.;5</i> will
return <i>a.</i> when VMS is (though that file could be opened with
<code class="inline"><a class="l_k" href="functions/open.html">open</a><span class="s">(</span><span class="w">FH</span><span class="cm">,</span> <span class="q">&#39;A&#39;</span><span class="s">)</span></code>
).</p>
<p>With support for extended file specifications and if <code class="inline"><a class="l_k" href="functions/opendir.html">opendir</a></code> was
given a UNIX format directory, a file named <i>A.;5</i> will return <i>a</i>
and optionally in the exact case on the disk.  When <code class="inline"><a class="l_k" href="functions/opendir.html">opendir</a></code> is given
a VMS format directory, then <code class="inline"><a class="l_k" href="functions/readdir.html">readdir</a></code> should return <i>a.</i>, and
again with the optionally the exact case.</p>
<p>RMS had an eight level limit on directory depths from any rooted logical
(allowing 16 levels overall) prior to VMS 7.2, and even with versions of
VMS on VAX up through 7.3.  Hence <code class="inline"><span class="j">PERL_ROOT:</span><span class="s">[</span><span class="w">LIB</span><span class="v">.2.3.4.5.6.7.8</span><span class="s">]</span></code>
 is a
valid directory specification but <code class="inline"><span class="j">PERL_ROOT:</span><span class="s">[</span><span class="w">LIB</span><span class="v">.2.3.4.5.6.7.8.9</span><span class="s">]</span></code>
 is
not.  <i>Makefile.PL</i> authors might have to take this into account, but at
least they can refer to the former as <code class="inline">/PERL_ROOT/lib/2/3/4/5/6/7/8/</code>.</p>
<p>Pumpkings and module integrators can easily see whether files with too many
directory levels have snuck into the core by running the following in the
top-level source directory:</p>
<pre class="verbatim"><ol><li>   <span class="i">$ perl</span> -ne <span class="q">&quot;$_=~s/\s+.*//; print if scalar(split /\//) &gt; 8;&quot;</span> &lt; <span class="w">MANIFEST</span></li></ol></pre><p>The VMS::Filespec module, which gets installed as part of the build
process on VMS, is a pure Perl module that can easily be installed on
non-VMS platforms and can be helpful for conversions to and from RMS
native formats.  It is also now the only way that you should check to
see if VMS is in a case sensitive mode.</p>
<p>What <code class="inline">\<span class="w">n</span></code>
 represents depends on the type of file opened.  It usually
represents <code class="inline">\<span class="n">012</span></code>
 but it could also be <code class="inline">\<span class="n">015</span></code>
, <code class="inline">\<span class="n">012</span></code>
, <code class="inline">\<span class="n">015</span>\<span class="n">012</span></code>
, 
<code class="inline">\<span class="n">000</span></code>
, <code class="inline">\<span class="n">040</span></code>
, or nothing depending on the file organization and 
record format.  The VMS::Stdio module provides access to the 
special fopen() requirements of files with unusual attributes on VMS.</p>
<p>TCP/IP stacks are optional on VMS, so socket routines might not be
implemented.  UDP sockets may not be supported.</p>
<p>The TCP/IP library support for all current versions of VMS is dynamically
loaded if present, so even if the routines are configured, they may
return a status indicating that they are not implemented.</p>
<p>The value of <code class="inline"><span class="i">$^O</span></code>
 on OpenVMS is &quot;VMS&quot;.  To determine the architecture
that you are running on without resorting to loading all of <code class="inline"><span class="i">%Config</span></code>

you can examine the content of the <code class="inline"><span class="i">@INC</span></code>
 array like so:</p>
<pre class="verbatim"><ol><li>    if <span class="s">(</span><a class="l_k" href="functions/grep.html">grep</a><span class="s">(</span><span class="q">/VMS_AXP/</span><span class="cm">,</span> <span class="i">@INC</span><span class="s">)</span><span class="s">)</span> <span class="s">{</span></li><li>        <a class="l_k" href="functions/print.html">print</a> <span class="q">&quot;I&#39;m on Alpha!\n&quot;</span><span class="sc">;</span></li><li></li><li>    <span class="s">}</span> elsif <span class="s">(</span><a class="l_k" href="functions/grep.html">grep</a><span class="s">(</span><span class="q">/VMS_VAX/</span><span class="cm">,</span> <span class="i">@INC</span><span class="s">)</span><span class="s">)</span> <span class="s">{</span></li><li>        <a class="l_k" href="functions/print.html">print</a> <span class="q">&quot;I&#39;m on VAX!\n&quot;</span><span class="sc">;</span></li><li></li><li>    <span class="s">}</span> elsif <span class="s">(</span><a class="l_k" href="functions/grep.html">grep</a><span class="s">(</span><span class="q">/VMS_IA64/</span><span class="cm">,</span> <span class="i">@INC</span><span class="s">)</span><span class="s">)</span> <span class="s">{</span></li><li>        <a class="l_k" href="functions/print.html">print</a> <span class="q">&quot;I&#39;m on IA64!\n&quot;</span><span class="sc">;</span></li><li></li><li>    <span class="s">}</span> else <span class="s">{</span></li><li>        <a class="l_k" href="functions/print.html">print</a> <span class="q">&quot;I&#39;m not so sure about where $^O is...\n&quot;</span><span class="sc">;</span></li><li>    <span class="s">}</span></li></ol></pre><p>In general, the significant differences should only be if Perl is running
on VMS_VAX or one of the 64 bit OpenVMS platforms.</p>
<p>On VMS, perl determines the UTC offset from the <code class="inline"><span class="w">SYS</span><span class="i">$TIMEZONE_DIFFERENTIAL</span></code>

logical name.  Although the VMS epoch began at 17-NOV-1858 00:00:00.00,
calls to <code class="inline"><a class="l_k" href="functions/localtime.html">localtime</a></code> are adjusted to count offsets from
01-JAN-1970 00:00:00.00, just like Unix.</p>
<p>Also see:</p>
<ul>
<li>
<p><i>README.vms</i> (installed as <a href="http://search.cpan.org/perldoc/README_vms">README_vms</a>), <a href="perlvms.html">perlvms</a></p>
</li>
<li>
<p>vmsperl list, vmsperl-subscribe@perl.org</p>
</li>
<li>
<p>vmsperl on the web, <a href="http://www.sidhe.org/vmsperl/index.html">http://www.sidhe.org/vmsperl/index.html</a></p>
</li>
</ul>
<a name="VOS"></a><h2>VOS</h2>
<p>Perl on VOS is discussed in <i>README.vos</i> in the perl distribution
(installed as <a href="perlvos.html">perlvos</a>).  Perl on VOS can accept either VOS- or
Unix-style file specifications as in either of the following:</p>
<pre class="verbatim"><ol><li>    $ perl -ne "print if /perl_setup/i" &gt;system&gt;notices</li><li>    $ perl -ne "print if /perl_setup/i" /system/notices</li></ol></pre><p>or even a mixture of both as in:</p>
<pre class="verbatim"><ol><li>    $ perl -ne "print if /perl_setup/i" &gt;system/notices</li></ol></pre><p>Even though VOS allows the slash character to appear in 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 names to 32 or fewer characters, file names cannot start with a
<code class="inline">-</code>
 character, or contain any character matching <code class="inline"><span class="q">tr/ !%&amp;'()*+;&lt;&gt;?//</span></code>
</p>
<p>The value of <code class="inline"><span class="i">$^O</span></code>
 on VOS is &quot;VOS&quot;.  To determine the architecture that
you are running on without resorting to loading all of <code class="inline"><span class="i">%Config</span></code>
 you
can examine the content of the @INC array like so:</p>
<pre class="verbatim"><ol><li>    if <span class="s">(</span><span class="i">$^O</span> =~ <span class="q">/VOS/</span><span class="s">)</span> <span class="s">{</span></li><li>        <a class="l_k" href="functions/print.html">print</a> <span class="q">&quot;I&#39;m on a Stratus box!\n&quot;</span><span class="sc">;</span></li><li>    <span class="s">}</span> else <span class="s">{</span></li><li>        <a class="l_k" href="functions/print.html">print</a> <span class="q">&quot;I&#39;m not on a Stratus box!\n&quot;</span><span class="sc">;</span></li><li>        <a class="l_k" href="functions/die.html">die</a><span class="sc">;</span></li><li>    <span class="s">}</span></li></ol></pre><p>Also see:</p>
<ul>
<li>
<p><i>README.vos</i> (installed as <a href="perlvos.html">perlvos</a>)</p>
</li>
<li>
<p>The VOS mailing list.</p>
<p>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 &quot;subscribe Info-Stratus&quot; in
the message body to majordomo@list.stratagy.com.</p>
</li>
<li>
<p>VOS Perl on the web at <a href="http://ftp.stratus.com/pub/vos/posix/posix.html">http://ftp.stratus.com/pub/vos/posix/posix.html</a></p>
</li>
</ul>
<a name="EBCDIC-Platforms"></a><h2>EBCDIC Platforms</h2>
<p>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 (usually
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 &quot;Unix system
services for OS/390&quot; (formerly known as OpenEdition), VM/ESA OpenEdition, or
the BS200 POSIX-BC system (BS2000 is supported in perl 5.6 and greater).
See <a href="perlos390.html">perlos390</a> 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 <a href="perlos400.html">perlos400</a>.</p>
<p>As of R2.5 of USS for OS/390 and Version 2.3 of VM/ESA these Unix
sub-systems do not support the <code class="inline"><span class="c">#!</span></code>
 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:</p>
<pre class="verbatim"><ol><li>    : # use perl</li><li>        <a class="l_k" href="functions/eval.html">eval</a> <span class="q">&#39;exec /usr/local/bin/perl -S $0 ${1+&quot;$@&quot;}&#39;</span></li><li>            if <span class="n">0</span><span class="sc">;</span></li><li>    <span class="c">#!/usr/local/bin/perl     # just a comment really</span></li><li></li><li>    <a class="l_k" href="functions/print.html">print</a> <span class="q">&quot;Hello from perl!\n&quot;</span><span class="sc">;</span></li></ol></pre><p>OS/390 will support the <code class="inline"><span class="c">#!</span></code>
 shebang trick in release 2.8 and beyond.
Calls to <code class="inline"><a class="l_k" href="functions/system.html">system</a></code> and backticks can use POSIX shell syntax on all
S/390 systems.</p>
<p>On the AS/400, if PERL5 is in your library list, you may need
to wrap your perl scripts in a CL procedure to invoke them like so:</p>
<pre class="verbatim"><ol><li>    BEGIN</li><li>      CALL PGM(PERL5/PERL) PARM('/QOpenSys/hello.pl')</li><li>    ENDPGM</li></ol></pre><p>This will invoke the perl script <i>hello.pl</i> in the root of the
QOpenSys file system.  On the AS/400 calls to <code class="inline"><a class="l_k" href="functions/system.html">system</a></code> or backticks
must use CL syntax.</p>
<p>On these platforms, bear in mind that the EBCDIC character set may have
an effect on what happens with some perl functions (such as <code class="inline"><a class="l_k" href="functions/chr.html">chr</a></code>,
<code class="inline"><a class="l_k" href="functions/pack.html">pack</a></code>, <code class="inline"><a class="l_k" href="functions/print.html">print</a></code>, <code class="inline"><a class="l_k" href="functions/printf.html">printf</a></code>, <code class="inline"><a class="l_k" href="functions/ord.html">ord</a></code>, <code class="inline"><a class="l_k" href="functions/sort.html">sort</a></code>, <code class="inline"><a class="l_k" href="functions/sprintf.html">sprintf</a></code>, <code class="inline"><a class="l_k" href="functions/unpack.html">unpack</a></code>), as
well as bit-fiddling with ASCII constants using operators like <code class="inline">^</code>, <code class="inline"><span class="i">&amp;</span></code>

and <code class="inline">|</code>, not to mention dealing with socket interfaces to ASCII computers
(see <a href="http://search.cpan.org/perldoc/&quot;Newlines&quot;">Newlines</a>).</p>
<p>Fortunately, most web servers for the mainframe will correctly
translate the <code class="inline">\<span class="w">n</span></code>
 in the following statement to its ASCII equivalent
(<code class="inline">\<span class="w">r</span></code>
 is the same under both Unix and OS/390 &amp; VM/ESA):</p>
<pre class="verbatim"><ol><li>    <a class="l_k" href="functions/print.html">print</a> <span class="q">&quot;Content-type: text/html\r\n\r\n&quot;</span><span class="sc">;</span></li></ol></pre><p>The values of <code class="inline"><span class="i">$^O</span></code>
 on some of these platforms includes:</p>
<pre class="verbatim"><ol><li>    uname         $^O        $Config{'archname'}</li><li>    --------------------------------------------</li><li>    OS/390        os390      os390</li><li>    OS400         os400      os400</li><li>    POSIX-BC      posix-bc   BS2000-posix-bc</li><li>    VM/ESA        vmesa      vmesa</li></ol></pre><p>Some simple tricks for determining if you are running on an EBCDIC
platform could include any of the following (perhaps all):</p>
<pre class="verbatim"><ol><li>    if <span class="s">(</span><span class="q">&quot;\t&quot;</span> eq <span class="q">&quot;\05&quot;</span><span class="s">)</span>   <span class="s">{</span> <a class="l_k" href="functions/print.html">print</a> <span class="q">&quot;EBCDIC may be spoken here!\n&quot;</span><span class="sc">;</span> <span class="s">}</span></li><li></li><li>    if <span class="s">(</span><a class="l_k" href="functions/ord.html">ord</a><span class="s">(</span><span class="q">&#39;A&#39;</span><span class="s">)</span> == <span class="n">193</span><span class="s">)</span> <span class="s">{</span> <a class="l_k" href="functions/print.html">print</a> <span class="q">&quot;EBCDIC may be spoken here!\n&quot;</span><span class="sc">;</span> <span class="s">}</span></li><li></li><li>    if <span class="s">(</span><a class="l_k" href="functions/chr.html">chr</a><span class="s">(</span><span class="n">169</span><span class="s">)</span> eq <span class="q">&#39;z&#39;</span><span class="s">)</span> <span class="s">{</span> <a class="l_k" href="functions/print.html">print</a> <span class="q">&quot;EBCDIC may be spoken here!\n&quot;</span><span class="sc">;</span> <span class="s">}</span></li></ol></pre><p>One thing you may not want to rely on is the EBCDIC encoding
of punctuation 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).</p>
<p>Also see:</p>
<ul>
<li>
<p><a href="perlos390.html">perlos390</a>, <i>README.os390</i>, <i>perlbs2000</i>, <i>README.vmesa</i>,
<a href="perlebcdic.html">perlebcdic</a>.</p>
</li>
<li>
<p>The perl-mvs@perl.org list is for discussion of porting issues as well as
general usage issues for all EBCDIC Perls.  Send a message body of
&quot;subscribe perl-mvs&quot; to majordomo@perl.org.</p>
</li>
<li>
<p>AS/400 Perl information at
<a href="http://as400.rochester.ibm.com/">http://as400.rochester.ibm.com/</a>
as well as on CPAN in the <i>ports/</i> directory.</p>
</li>
</ul>
<a name="Acorn-RISC-OS"></a><h2>Acorn RISC OS</h2>
<p>Because Acorns use ASCII with newlines (<code class="inline">\<span class="w">n</span></code>
) in text files as <code class="inline">\<span class="n">012</span></code>
 like
Unix, and because Unix filename emulation is turned on by default, 
most simple scripts will probably work &quot;out of the box&quot;.  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 and directory
names are silently truncated to fit.  Scripts should be aware that the
standard filesystem currently has a name length limit of <b>10</b>
characters, with up to 77 items in a directory, but other filesystems
may not impose such limitations.</p>
<p>Native filenames are of the form</p>
<pre class="verbatim"><ol><li>    <span class="w">Filesystem</span><span class="c">#Special_Field::DiskName.$.Directory.Directory.File</span></li></ol></pre><p>where</p>
<pre class="verbatim"><ol><li>    Special_Field is not usually present, but may contain . and $ .</li><li>    Filesystem =~ m|[A-Za-z0-9_]|</li><li>    DsicName   =~ m|[A-Za-z0-9_/]|</li><li>    $ represents the root directory</li><li>    . is the path separator</li><li>    @ is the current directory (per filesystem but machine global)</li><li>    ^ is the parent directory</li><li>    Directory and File =~ m|[^\0- "\.\$\%\&amp;:\@\\^\|\177]+|</li></ol></pre><p>The default filename translation is roughly <code class="inline"><a class="l_k" href="functions/tr.html">tr|/.|./|;</a></code></p>
<p>Note that <code class="inline"><span class="q">&quot;ADFS::HardDisk.$.File&quot;</span> ne <span class="q">&#39;ADFS::HardDisk.$.File&#39;</span></code>
 and that
the second stage of <code class="inline"><span class="i">$</span></code>
 interpolation in regular expressions will fall
foul of the <code class="inline"><span class="i">$.</span></code>
 if scripts are not careful.</p>
<p>Logical paths specified by system variables containing comma-separated
search lists are also allowed; hence <code class="inline"><span class="j">System:</span><span class="w">Modules</span></code>
 is a valid
filename, and the filesystem will prefix <code class="inline"><span class="w">Modules</span></code>
 with each section of
<code class="inline"><span class="w">System</span><span class="i">$Path</span></code>
 until a name is made that points to an object on disk.
Writing to a new file <code class="inline"><span class="j">System:</span><span class="w">Modules</span></code>
 would be allowed only if
<code class="inline"><span class="w">System</span><span class="i">$Path</span></code>
 contains a single item list.  The filesystem will also
expand system variables in filenames if enclosed in angle brackets, so
<code class="inline"><span class="q">&lt;System$Dir&gt;</span>.<span class="w">Modules</span></code>
 would look for the file
<code class="inline"><span class="i">$ENV</span>{<span class="q">&#39;System$Dir&#39;</span>} . <span class="q">&#39;Modules&#39;</span></code>
.  The obvious implication of this is
that <b>fully qualified filenames can start with <code class="inline">&lt;&gt;&gt;&gt;&gt; and should
be protected when &lt;code class="inline"&gt;&lt;a class="l_k" href="functions/open.html"&gt;open&lt;/a&gt;&lt;/code&gt; is used for input.</code></b></p>
<p>Because <code class="inline">.</code> was in 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 <code class="inline">.c</code> <code class="inline">.h</code> <code class="inline">.s</code> and <code class="inline">.o</code> suffix from
filenames specified in source code and store the respective files in
subdirectories named after the suffix.  Hence files are translated:</p>
<pre class="verbatim"><ol><li>    foo.h           h.foo</li><li>    C:foo.h         C:h.foo        (logical path variable)</li><li>    sys/os.h        sys.h.os       (C compiler groks Unix-speak)</li><li>    10charname.c    c.10charname</li><li>    10charname.o    o.10charname</li><li>    11charname_.c   c.11charname   (assuming filesystem truncates at 10)</li></ol></pre><p>The Unix emulation library&#39;s translation of filenames to native assumes
that this sort of translation is required, and it allows a user-defined list
of known suffixes that it will transpose in this fashion.  This may
seem transparent, but consider that with these rules <code class="inline"><span class="w">foo</span>/<span class="w">bar</span>/<span class="w">baz</span>.<span class="w">h</span></code>

and <code class="inline"><span class="w">foo</span>/<span class="w">bar</span>/<span class="w">h</span>/<span class="w">baz</span></code>
 both map to <code class="inline"><span class="w">foo</span>.<span class="w">bar</span>.<span class="w">h</span>.<span class="w">baz</span></code>
, and that <code class="inline"><a class="l_k" href="functions/readdir.html">readdir</a></code> and
<code class="inline"><a class="l_k" href="functions/glob.html">glob</a></code> cannot and do not attempt to emulate the reverse mapping.  Other
<code class="inline">.</code>&#39;s in filenames are translated to <code class="inline">/</code>.</p>
<p>As implied above, the environment accessed through <code class="inline"><span class="i">%ENV</span></code>
 is global, and
the convention is that program specific environment variables are of the
form <code class="inline"><span class="w">Program</span><span class="i">$Name</span></code>
.  Each filesystem maintains a current directory,
and the current filesystem&#39;s current directory is the <b>global</b> current
directory.  Consequently, sociable programs don&#39;t change the current
directory but rely on full pathnames, and programs (and Makefiles) cannot
assume that they can spawn a child process which can change the current
directory without affecting its parent (and everyone else for that
matter).</p>
<p>Because native operating system filehandles are global and are currently 
allocated down from 255, with 0 being a reserved value, the Unix emulation
library emulates Unix filehandles.  Consequently, you can&#39;t rely on
passing <code class="inline"><span class="w">STDIN</span></code>
, <code class="inline"><span class="w">STDOUT</span></code>
, or <code class="inline"><span class="w">STDERR</span></code>
 to your children.</p>
<p>The desire of users to express filenames of the form
<code class="inline"><span class="q">&lt;Foo$Dir&gt;</span>.<span class="w">Bar</span></code>
 on the command line unquoted causes problems,
too: <code class="inline"><span class="q">``</span></code>
 command output capture has to perform a guessing game.  It
assumes that a string <code class="inline">&lt;[^&lt;&gt;]+\$[^&lt;&gt;]&gt;</code> is a
reference to an environment variable, whereas anything else involving
<code class="inline">&lt;</code>
 or <code class="inline">&gt;</code> is redirection, and generally manages to be 99%
right.  Of course, the problem remains that scripts cannot rely on any
Unix tools being available, or that any tools found have Unix-like command
line arguments.</p>
<p>Extensions and XS are, in theory, buildable by anyone using free
tools.  In practice, many don&#39;t, as users of the Acorn platform are
used to binary distributions.  MakeMaker does run, but no available
make currently copes with MakeMaker&#39;s makefiles; even if and when
this should be fixed, the lack of a Unix-like shell will cause
problems with makefile rules, especially lines of the form <code class="inline"><span class="w">cd</span>
<span class="w">sdbm</span> &amp;&amp; <span class="w">make</span> <span class="w">all</span></code>
, and anything using quoting.</p>
<p>&quot;RISC OS&quot; is the proper name for the operating system, but the value
in <code class="inline"><span class="i">$^O</span></code>
 is &quot;riscos&quot; (because we don&#39;t like shouting).</p>
<a name="Other-perls"></a><h2>Other perls</h2>
<p>Perl has been ported to many platforms that do not fit into any of
the categories listed above.  Some, such as AmigaOS, Atari MiNT,
BeOS, HP MPE/iX, QNX, Plan 9, and VOS, have been well-integrated
into the standard Perl source code kit.  You may need to see the
<i>ports/</i> directory on CPAN for information, and possibly binaries,
for the likes of: aos, Atari ST, lynxos, riscos, Novell Netware,
Tandem Guardian, <i>etc.</i>  (Yes, we know that some of these OSes may
fall under the Unix category, but we are not a standards body.)</p>
<p>Some approximate operating system names and their <code class="inline"><span class="i">$^O</span></code>
 values
in the &quot;OTHER&quot; category include:</p>
<pre class="verbatim"><ol><li>    OS            $^O        $Config{'archname'}</li><li>    ------------------------------------------</li><li>    Amiga DOS     amigaos    m68k-amigos</li><li>    BeOS          beos</li><li>    MPE/iX        mpeix      PA-RISC1.1</li></ol></pre><p>See also:</p>
<ul>
<li>
<p>Amiga, <i>README.amiga</i> (installed as <a href="perlamiga.html">perlamiga</a>).</p>
</li>
<li>
<p>Atari, <i>README.mint</i> and Guido Flohr&#39;s web page
<a href="http://stud.uni-sb.de/~gufl0000/">http://stud.uni-sb.de/~gufl0000/</a></p>
</li>
<li>
<p>Be OS, <i>README.beos</i></p>
</li>
<li>
<p>HP 300 MPE/iX, <i>README.mpeix</i> and Mark Bixby&#39;s web page
<a href="http://www.bixby.org/mark/perlix.html">http://www.bixby.org/mark/perlix.html</a></p>
</li>
<li>
<p>A free perl5-based PERL.NLM for Novell Netware is available in
precompiled binary and source code form from <a href="http://www.novell.com/">http://www.novell.com/</a>
as well as from CPAN.</p>
</li>
<li>
<p>Plan 9, <i>README.plan9</i></p>
</li>
</ul>
<a name="FUNCTION-IMPLEMENTATIONS"></a><h1>FUNCTION IMPLEMENTATIONS</h1>
<p>Listed below are functions that are either completely unimplemented
or else have been implemented differently on various platforms.
Following each description will be, in parentheses, a list of
platforms that the description applies to.</p>
<p>The list may well be incomplete, or even wrong in some places.  When
in doubt, consult the platform-specific README files in the Perl
source distribution, and any other documentation resources accompanying
a given port.</p>
<p>Be aware, moreover, that even among Unix-ish systems there are variations.</p>
<p>For many functions, you can also query <code class="inline"><span class="i">%Config</span></code>
, exported by
default from the Config module.  For example, to check whether the
platform has the <code class="inline"><a class="l_k" href="functions/lstat.html">lstat</a></code> call, check <code class="inline"><span class="i">$Config</span>{<span class="w">d_lstat</span>}</code>
.  See
<a href="Config.html">Config</a> for a full description of available variables.</p>
<a name="Alphabetical-Listing-of-Perl-Functions"></a><h2>Alphabetical Listing of Perl Functions</h2>
<ul>
<li><a name="-X"></a><b>-X</b>
<p><code class="inline">-r</code>
, <code class="inline">-w</code>
, and <code class="inline">-x</code>
 have a limited meaning only; directories
and applications are executable, and there are no uid/gid
considerations.  <code class="inline">-o</code>
 is not supported.  (Mac OS)</p>
<p><code class="inline">-w</code>
 only inspects the read-only file attribute (FILE_ATTRIBUTE_READONLY),
which determines whether the directory can be deleted, not whether it can
be written to. Directories always have read and write access unless denied
by discretionary access control lists (DACLs).  (Win32)</p>
<p><code class="inline">-r</code>
, <code class="inline">-w</code>
, <code class="inline">-x</code>
, and <code class="inline">-o</code>
 tell whether the file is accessible,
which may not reflect UIC-based file protections.  (VMS)</p>
<p><code class="inline">-s</code>
 returns the size of the data fork, not the total size of data fork
plus resource fork.  (Mac OS).</p>
<p><code class="inline">-s</code>
 by name on an open file will return the space reserved on disk,
rather than the current extent.  <code class="inline">-s</code>
 on an open filehandle returns the
current size.  (RISC OS)</p>
<p><code class="inline">-R</code>
, <code class="inline">-W</code>
, <code class="inline"><a class="l_k" href="functions/-X.html">-X</a></code>, <code class="inline">-O</code>
 are indistinguishable from <code class="inline">-r</code>
, <code class="inline">-w</code>
,
<code class="inline">-x</code>
, <code class="inline">-o</code>
. (Mac OS, Win32, VMS, RISC OS)</p>
<p><code class="inline">-b</code>
, <code class="inline">-c</code>
, <code class="inline">-k</code>
, <code class="inline">-g</code>
, <code class="inline">-p</code>
, <code class="inline">-u</code>
, <code class="inline">-A</code>
 are not implemented.
(Mac OS)</p>
<p><code class="inline">-g</code>
, <code class="inline">-k</code>
, <code class="inline">-l</code>
, <code class="inline">-p</code>
, <code class="inline">-u</code>
, <code class="inline">-A</code>
 are not particularly meaningful.
(Win32, VMS, RISC OS)</p>
<p><code class="inline">-d</code>
 is true if passed a device spec without an explicit directory.
(VMS)</p>
<p><code class="inline">-T</code>
 and <code class="inline">-B</code>
 are implemented, but might misclassify Mac text files
with foreign characters; this is the case will all platforms, but may
affect Mac OS often.  (Mac OS)</p>
<p><code class="inline">-x</code>
 (or <code class="inline"><a class="l_k" href="functions/-X.html">-X</a></code>) determine if a file ends in one of the executable
suffixes.  <code class="inline">-S</code>
 is meaningless.  (Win32)</p>
<p><code class="inline">-x</code>
 (or <code class="inline"><a class="l_k" href="functions/-X.html">-X</a></code>) determine if a file has an executable file type.
(RISC OS)</p>
</li>
<li><a name="atan2"></a><b>atan2</b>
<p>Due to issues with various CPUs, math libraries, compilers, and standards,
results for <code class="inline"><a class="l_k" href="functions/atan2.html">atan2()</a></code> may vary depending on any combination of the above.
Perl attempts to conform to the Open Group/IEEE standards for the results
returned from <code class="inline"><a class="l_k" href="functions/atan2.html">atan2()</a></code>, but cannot force the issue if the system Perl is
run on does not allow it.  (Tru64, HP-UX 10.20)</p>
<p>The current version of the standards for <code class="inline"><a class="l_k" href="functions/atan2.html">atan2()</a></code> is available at 
<a href="http://www.opengroup.org/onlinepubs/009695399/functions/atan2.html">http://www.opengroup.org/onlinepubs/009695399/functions/atan2.html</a>.</p>
</li>
<li><a name="binmode"></a><b>binmode</b>
<p>Meaningless.  (Mac OS, RISC OS)</p>
<p>Reopens file and restores pointer; if function fails, underlying
filehandle may be closed, or pointer may be in a different position.
(VMS)</p>
<p>The value returned by <code class="inline"><a class="l_k" href="functions/tell.html">tell</a></code> may be affected after the call, and
the filehandle may be flushed. (Win32)</p>
</li>
<li><a name="chmod"></a><b>chmod</b>
<p>Only limited meaning.  Disabling/enabling write permission is mapped to
locking/unlocking the file. (Mac OS)</p>
<p>Only good for changing &quot;owner&quot; read-write access, &quot;group&quot;, and &quot;other&quot;
bits are meaningless. (Win32)</p>
<p>Only good for changing &quot;owner&quot; and &quot;other&quot; read-write access. (RISC OS)</p>
<p>Access permissions are mapped onto VOS access-control list changes. (VOS)</p>
<p>The actual permissions set depend on the value of the <code class="inline"><span class="w">CYGWIN</span></code>

in the SYSTEM environment settings.  (Cygwin)</p>
</li>
<li><a name="chown"></a><b>chown</b>
<p>Not implemented. (Mac OS, Win32, Plan 9, RISC OS)</p>
<p>Does nothing, but won&#39;t fail. (Win32)</p>
<p>A little funky, because VOS&#39;s notion of ownership is a little funky (VOS).</p>
</li>
<li><a name="chroot"></a><b>chroot</b>
<p>Not implemented. (Mac OS, Win32, VMS, Plan 9, RISC OS, VOS, VM/ESA)</p>
</li>
<li><a name="crypt"></a><b>crypt</b>
<p>May not be available if library or source was not provided when building
perl. (Win32)</p>
</li>
<li><a name="dbmclose"></a><b>dbmclose</b>
<p>Not implemented. (VMS, Plan 9, VOS)</p>
</li>
<li><a name="dbmopen"></a><b>dbmopen</b>
<p>Not implemented. (VMS, Plan 9, VOS)</p>
</li>
<li><a name="dump"></a><b>dump</b>
<p>Not useful. (Mac OS, RISC OS)</p>
<p>Not supported. (Cygwin, Win32)</p>
<p>Invokes VMS debugger. (VMS)</p>
</li>
<li><a name="exec"></a><b>exec</b>
<p>Not implemented. (Mac OS)</p>
<p>Implemented via Spawn. (VM/ESA)</p>
<p>Does not automatically flush output handles on some platforms.
(SunOS, Solaris, HP-UX)</p>
</li>
<li><a name="exit"></a><b>exit</b>
<p>Emulates UNIX exit() (which considers <code class="inline"><a class="l_k" href="functions/exit.html">exit</a> <span class="n">1</span></code>
 to indicate an error) by
mapping the <code class="inline"><span class="n">1</span></code>
 to SS$_ABORT (<code class="inline"><span class="n">44</span></code>
).  This behavior may be overridden
with the pragma <code class="inline"><a class="l_k" href="functions/use.html">use</a> <span class="w">vmsish</span> <span class="q">&#39;exit&#39;</span></code>
.  As with the CRTL&#39;s exit()
function, <code class="inline"><a class="l_k" href="functions/exit.html">exit</a> <span class="n">0</span></code>
 is also mapped to an exit status of SS$_NORMAL
(<code class="inline"><span class="n">1</span></code>
); this mapping cannot be overridden.  Any other argument to exit()
is used directly as Perl&#39;s exit status.  On VMS, unless the future
POSIX_EXIT mode is enabled, the exit code should always be a valid
VMS exit code and not a generic number.  When the POSIX_EXIT mode is
enabled, a generic number will be encoded in a method compatible with
the C library _POSIX_EXIT macro so that it can be decoded by other
programs, particularly ones written in C, like the GNV package.  (VMS)</p>
</li>
<li><a name="fcntl"></a><b>fcntl</b>
<p>Not implemented. (Win32)
Some functions available based on the version of VMS. (VMS)</p>
</li>
<li><a name="flock"></a><b>flock</b>
<p>Not implemented (Mac OS, VMS, RISC OS, VOS).</p>
<p>Available only on Windows NT (not on Windows 95). (Win32)</p>
</li>
<li><a name="fork"></a><b>fork</b>
<p>Not implemented. (Mac OS, AmigaOS, RISC OS, VM/ESA, VMS)</p>
<p>Emulated using multiple interpreters.  See <a href="perlfork.html">perlfork</a>.  (Win32)</p>
<p>Does not automatically flush output handles on some platforms.
(SunOS, Solaris, HP-UX)</p>
</li>
<li><a name="getlogin"></a><b>getlogin</b>
<p>Not implemented. (Mac OS, RISC OS)</p>
</li>
<li><a name="getpgrp"></a><b>getpgrp</b>
<p>Not implemented. (Mac OS, Win32, VMS, RISC OS)</p>
</li>
<li><a name="getppid"></a><b>getppid</b>
<p>Not implemented. (Mac OS, Win32, RISC OS)</p>
</li>
<li><a name="getpriority"></a><b>getpriority</b>
<p>Not implemented. (Mac OS, Win32, VMS, RISC OS, VOS, VM/ESA)</p>
</li>
<li><a name="getpwnam"></a><b>getpwnam</b>
<p>Not implemented. (Mac OS, Win32)</p>
<p>Not useful. (RISC OS)</p>
</li>
<li><a name="getgrnam"></a><b>getgrnam</b>
<p>Not implemented. (Mac OS, Win32, VMS, RISC OS)</p>
</li>
<li><a name="getnetbyname"></a><b>getnetbyname</b>
<p>Not implemented. (Mac OS, Win32, Plan 9)</p>
</li>
<li><a name="getpwuid"></a><b>getpwuid</b>
<p>Not implemented. (Mac OS, Win32)</p>
<p>Not useful. (RISC OS)</p>
</li>
<li><a name="getgrgid"></a><b>getgrgid</b>
<p>Not implemented. (Mac OS, Win32, VMS, RISC OS)</p>
</li>
<li><a name="getnetbyaddr"></a><b>getnetbyaddr</b>
<p>Not implemented. (Mac OS, Win32, Plan 9)</p>
</li>
<li><a name="getprotobynumber"></a><b>getprotobynumber</b>
<p>Not implemented. (Mac OS)</p>
</li>
<li><a name="getservbyport"></a><b>getservbyport</b>
<p>Not implemented. (Mac OS)</p>
</li>
<li><a name="getpwent"></a><b>getpwent</b>
<p>Not implemented. (Mac OS, Win32, VM/ESA)</p>
</li>
<li><a name="getgrent"></a><b>getgrent</b>
<p>Not implemented. (Mac OS, Win32, VMS, VM/ESA)</p>
</li>
<li><a name="gethostbyname"></a><b>gethostbyname</b>
<p><code class="inline"><a class="l_k" href="functions/gethostbyname.html">gethostbyname('localhost')</a></code> does not work everywhere: you may have
to use <code class="inline"><a class="l_k" href="functions/gethostbyname.html">gethostbyname('127.0.0.1')</a></code>. (Mac OS, Irix 5)</p>
</li>
<li><a name="gethostent"></a><b>gethostent</b>
<p>Not implemented. (Mac OS, Win32)</p>
</li>
<li><a name="getnetent"></a><b>getnetent</b>
<p>Not implemented. (Mac OS, Win32, Plan 9)</p>
</li>
<li><a name="getprotoent"></a><b>getprotoent</b>
<p>Not implemented. (Mac OS, Win32, Plan 9)</p>
</li>
<li><a name="getservent"></a><b>getservent</b>
<p>Not implemented. (Win32, Plan 9)</p>
</li>
<li><a name="sethostent"></a><b>sethostent</b>
<p>Not implemented. (Mac OS, Win32, Plan 9, RISC OS)</p>
</li>
<li><a name="setnetent"></a><b>setnetent</b>
<p>Not implemented. (Mac OS, Win32, Plan 9, RISC OS)</p>
</li>
<li><a name="setprotoent"></a><b>setprotoent</b>
<p>Not implemented. (Mac OS, Win32, Plan 9, RISC OS)</p>
</li>
<li><a name="setservent"></a><b>setservent</b>
<p>Not implemented. (Plan 9, Win32, RISC OS)</p>
</li>
<li><a name="endpwent"></a><b>endpwent</b>
<p>Not implemented. (Mac OS, MPE/iX, VM/ESA, Win32)</p>
</li>
<li><a name="endgrent"></a><b>endgrent</b>
<p>Not implemented. (Mac OS, MPE/iX, RISC OS, VM/ESA, VMS, Win32)</p>
</li>
<li><a name="endhostent"></a><b>endhostent</b>
<p>Not implemented. (Mac OS, Win32)</p>
</li>
<li><a name="endnetent"></a><b>endnetent</b>
<p>Not implemented. (Mac OS, Win32, Plan 9)</p>
</li>
<li><a name="endprotoent"></a><b>endprotoent</b>
<p>Not implemented. (Mac OS, Win32, Plan 9)</p>
</li>
<li><a name="endservent"></a><b>endservent</b>
<p>Not implemented. (Plan 9, Win32)</p>
</li>
<li><a name="getsockopt-SOCKET%2cLEVEL%2cOPTNAME"></a><b>getsockopt SOCKET,LEVEL,OPTNAME</b>
<p>Not implemented. (Plan 9)</p>
</li>
<li><a name="glob"></a><b>glob</b>
<p>This operator is implemented via the File::Glob extension on most
platforms.  See <a href="File/Glob.html">File::Glob</a> for portability information.</p>
</li>
<li><a name="gmtime"></a><b>gmtime</b>
<p>Same portability caveats as <a href="functions/localtime.html">localtime</a>.</p>
</li>
<li><a name="ioctl-FILEHANDLE%2cFUNCTION%2cSCALAR"></a><b>ioctl FILEHANDLE,FUNCTION,SCALAR</b>
<p>Not implemented. (VMS)</p>
<p>Available only for socket handles, and it does what the ioctlsocket() call
in the Winsock API does. (Win32)</p>
<p>Available only for socket handles. (RISC OS)</p>
</li>
<li><a name="kill"></a><b>kill</b>
<p><code class="inline"><a class="l_k" href="functions/kill.html">kill</a><span class="s">(</span><span class="n">0</span><span class="cm">,</span> <span class="w">LIST</span><span class="s">)</span></code>
 is implemented for the sake of taint checking;
use with other signals is unimplemented. (Mac OS)</p>
<p>Not implemented, hence not useful for taint checking. (RISC OS)</p>
<p><code class="inline"><a class="l_k" href="functions/kill.html">kill()</a></code> doesn&#39;t have the semantics of <code class="inline"><span class="i">raise</span><span class="s">(</span><span class="s">)</span></code>
, i.e. it doesn&#39;t send
a signal to the identified process like it does on Unix platforms.
Instead <code class="inline"><a class="l_k" href="functions/kill.html">kill</a><span class="s">(</span><span class="i">$sig</span><span class="cm">,</span> <span class="i">$pid</span><span class="s">)</span></code>
 terminates the process identified by $pid,
and makes it exit immediately with exit status $sig.  As in Unix, if
$sig is 0 and the specified process exists, it returns true without
actually terminating it. (Win32)</p>
<p><code class="inline"><a class="l_k" href="functions/kill.html">kill</a><span class="s">(</span><span class="n">-9</span><span class="cm">,</span> <span class="i">$pid</span><span class="s">)</span></code>
 will terminate the process specified by $pid and
recursively all child processes owned by it.  This is different from
the Unix semantics, where the signal will be delivered to all
processes in the same process group as the process specified by
$pid. (Win32)</p>
<p>Is not supported for process identification number of 0 or negative
numbers. (VMS)</p>
</li>
<li><a name="link"></a><b>link</b>
<p>Not implemented. (Mac OS, MPE/iX, RISC OS)</p>
<p>Link count not updated because hard links are not quite that hard
(They are sort of half-way between hard and soft links). (AmigaOS)</p>
<p>Hard links are implemented on Win32 under NTFS only. They are
natively supported on Windows 2000 and later.  On Windows NT they
are implemented using the Windows POSIX subsystem support and the
Perl process will need Administrator or Backup Operator privileges
to create hard links.</p>
<p>Available on 64 bit OpenVMS 8.2 and later.  (VMS)</p>
</li>
<li><a name="localtime"></a><b>localtime</b>
<p>Because Perl currently relies on the native standard C localtime()
function, it is only safe to use times between 0 and (2**31)-1.  Times
outside this range may result in unexpected behavior depending on your
operating system&#39;s implementation of localtime().</p>
</li>
<li><a name="lstat"></a><b>lstat</b>
<p>Not implemented. (RISC OS)</p>
<p>Return values (especially for device and inode) may be bogus. (Win32)</p>
</li>
<li><a name="msgctl"></a><b>msgctl</b>
</li>
<li><a name="msgget"></a><b>msgget</b>
</li>
<li><a name="msgsnd"></a><b>msgsnd</b>
</li>
<li><a name="msgrcv"></a><b>msgrcv</b>
<p>Not implemented. (Mac OS, Win32, VMS, Plan 9, RISC OS, VOS)</p>
</li>
<li><a name="open"></a><b>open</b>
<p>The <code class="inline">|</code> variants are supported only if ToolServer is installed.
(Mac OS)</p>
<p>open to <code class="inline">|-</code> and <code class="inline">-|</code> are unsupported. (Mac OS, Win32, RISC OS)</p>
<p>Opening a process does not automatically flush output handles on some
platforms.  (SunOS, Solaris, HP-UX)</p>
</li>
<li><a name="pipe"></a><b>pipe</b>
<p>Very limited functionality. (MiNT)</p>
</li>
<li><a name="readlink"></a><b>readlink</b>
<p>Not implemented. (Win32, VMS, RISC OS)</p>
</li>
<li><a name="rename"></a><b>rename</b>
<p>Can&#39;t move directories between directories on different logical volumes. (Win32)</p>
</li>
<li><a name="select"></a><b>select</b>
<p>Only implemented on sockets. (Win32, VMS)</p>
<p>Only reliable on sockets. (RISC OS)</p>
<p>Note that the <code class="inline"><a class="l_k" href="functions/select.html">select</a> <span class="w">FILEHANDLE</span></code>
 form is generally portable.</p>
</li>
<li><a name="semctl"></a><b>semctl</b>
</li>
<li><a name="semget"></a><b>semget</b>
</li>
<li><a name="semop"></a><b>semop</b>
<p>Not implemented. (Mac OS, Win32, VMS, RISC OS, VOS)</p>
</li>
<li><a name="setgrent"></a><b>setgrent</b>
<p>Not implemented. (Mac OS, MPE/iX, VMS, Win32, RISC OS, VOS)</p>
</li>
<li><a name="setpgrp"></a><b>setpgrp</b>
<p>Not implemented. (Mac OS, Win32, VMS, RISC OS, VOS)</p>
</li>
<li><a name="setpriority"></a><b>setpriority</b>
<p>Not implemented. (Mac OS, Win32, VMS, RISC OS, VOS)</p>
</li>
<li><a name="setpwent"></a><b>setpwent</b>
<p>Not implemented. (Mac OS, MPE/iX, Win32, RISC OS, VOS)</p>
</li>
<li><a name="setsockopt"></a><b>setsockopt</b>
<p>Not implemented. (Plan 9)</p>
</li>
<li><a name="shmctl"></a><b>shmctl</b>
</li>
<li><a name="shmget"></a><b>shmget</b>
</li>
<li><a name="shmread"></a><b>shmread</b>
</li>
<li><a name="shmwrite"></a><b>shmwrite</b>
<p>Not implemented. (Mac OS, Win32, VMS, RISC OS, VOS)</p>
</li>
<li><a name="sockatmark"></a><b>sockatmark</b>
<p>A relatively recent addition to socket functions, may not
be implemented even in UNIX platforms.</p>
</li>
<li><a name="socketpair"></a><b>socketpair</b>
<p>Not implemented. (RISC OS, VOS, VM/ESA)</p>
<p>Available on 64 bit OpenVMS 8.2 and later.  (VMS)</p>
</li>
<li><a name="stat"></a><b>stat</b>
<p>Platforms that do not have rdev, blksize, or blocks will return these
as &#39;&#39;, so numeric comparison or manipulation of these fields may cause
&#39;not numeric&#39; warnings.</p>
<p>mtime and atime are the same thing, and ctime is creation time instead of
inode change time. (Mac OS).</p>
<p>ctime not supported on UFS (Mac OS X).</p>
<p>ctime is creation time instead of inode change time  (Win32).</p>
<p>device and inode are not meaningful.  (Win32)</p>
<p>device and inode are not necessarily reliable.  (VMS)</p>
<p>mtime, atime and ctime all return the last modification time.  Device and
inode are not necessarily reliable.  (RISC OS)</p>
<p>dev, rdev, blksize, and blocks are not available.  inode is not
meaningful and will differ between stat calls on the same file.  (os2)</p>
<p>some versions of cygwin when doing a stat(&quot;foo&quot;) and if not finding it
may then attempt to stat(&quot;foo.exe&quot;) (Cygwin)</p>
<p>On Win32 stat() needs to open the file to determine the link count
and update attributes that may have been changed through hard links.
Setting ${^WIN32_SLOPPY_STAT} to a true value speeds up stat() by
not performing this operation. (Win32)</p>
</li>
<li><a name="symlink"></a><b>symlink</b>
<p>Not implemented. (Win32, RISC OS)</p>
<p>Implemented on 64 bit VMS 8.3.  VMS requires the symbolic link to be in Unix
syntax if it is intended to resolve to a valid path.</p>
</li>
<li><a name="syscall"></a><b>syscall</b>
<p>Not implemented. (Mac OS, Win32, VMS, RISC OS, VOS, VM/ESA)</p>
</li>
<li><a name="sysopen"></a><b>sysopen</b>
<p>The traditional &quot;0&quot;, &quot;1&quot;, and &quot;2&quot; MODEs are implemented with different
numeric values on some systems.  The flags exported by <code class="inline"><span class="w">Fcntl</span></code>

(O_RDONLY, O_WRONLY, O_RDWR) should work everywhere though.  (Mac
OS, OS/390, VM/ESA)</p>
</li>
<li><a name="system"></a><b>system</b>
<p>Only implemented if ToolServer is installed. (Mac OS)</p>
<p>As an optimization, may not call the command shell specified in
<code class="inline"><span class="i">$ENV</span>{<span class="w">PERL5SHELL</span>}</code>
.  <code class="inline"><a class="l_k" href="functions/system.html">system</a><span class="s">(</span><span class="n">1</span><span class="cm">,</span> <span class="i">@args</span><span class="s">)</span></code>
 spawns an external
process and immediately returns its process designator, without
waiting for it to terminate.  Return value may be used subsequently
in <code class="inline"><a class="l_k" href="functions/wait.html">wait</a></code> or <code class="inline"><a class="l_k" href="functions/waitpid.html">waitpid</a></code>.  Failure to spawn() a subprocess is indicated
by setting $? to &quot;255 &lt;&lt; 8&quot;.  <code class="inline"><span class="i">$?</span></code>
 is set in a way compatible with
Unix (i.e. the exitstatus of the subprocess is obtained by &quot;$?&gt;&gt; 8&quot;,
as described in the documentation).  (Win32)</p>
<p>There is no shell to process metacharacters, and the native standard is
to pass a command line terminated by &quot;\n&quot; &quot;\r&quot; or &quot;\0&quot; to the spawned
program.  Redirection such as <code class="inline">&gt; foo</code> is performed (if at all) by
the run time library of the spawned program.  <code class="inline"><a class="l_k" href="functions/system.html">system</a></code> <i>list</i> will call
the Unix emulation library&#39;s <code class="inline"><a class="l_k" href="functions/exec.html">exec</a></code> emulation, which attempts to provide
emulation of the stdin, stdout, stderr in force in the parent, providing
the child program uses a compatible version of the emulation library.
<i>scalar</i> will call the native command line direct and no such emulation
of a child Unix program will exists.  Mileage <b>will</b> vary.  (RISC OS)</p>
<p>Far from being POSIX compliant.  Because there may be no underlying
/bin/sh tries to work around the problem by forking and execing the
first token in its argument string.  Handles basic redirection
(&quot;&lt;&quot; or &quot;&gt;&quot;) on its own behalf. (MiNT)</p>
<p>Does not automatically flush output handles on some platforms.
(SunOS, Solaris, HP-UX)</p>
<p>The return value is POSIX-like (shifted up by 8 bits), which only allows
room for a made-up value derived from the severity bits of the native
32-bit condition code (unless overridden by <code class="inline"><a class="l_k" href="functions/use.html">use</a> <span class="w">vmsish</span> <span class="q">&#39;status&#39;</span></code>
). 
If the native condition code is one that has a POSIX value encoded, the
POSIX value will be decoded to extract the expected exit value.
For more details see <a href="perlvms.html#%24%3f">"$?" in perlvms</a>. (VMS)</p>
</li>
<li><a name="times"></a><b>times</b>
<p>Only the first entry returned is nonzero. (Mac OS)</p>
<p>&quot;cumulative&quot; times will be bogus.  On anything other than Windows NT
or Windows 2000, &quot;system&quot; time will be bogus, and &quot;user&quot; time is
actually the time returned by the clock() function in the C runtime
library. (Win32)</p>
<p>Not useful. (RISC OS)</p>
</li>
<li><a name="truncate"></a><b>truncate</b>
<p>Not implemented. (Older versions of VMS)</p>
<p>Truncation to same-or-shorter lengths only. (VOS)</p>
<p>If a FILEHANDLE is supplied, it must be writable and opened in append
mode (i.e., use <code class="inline"><a class="l_k" href="functions/open.html">open</a><span class="s">(</span><span class="w">FH</span><span class="cm">,</span> <span class="q">&#39;&gt;&gt;filename&#39;</span><span class="s">)</span></code>

or <code class="inline"><a class="l_k" href="functions/sysopen.html">sysopen(FH,...,O_APPEND|O_RDWR)</a></code>.  If a filename is supplied, it
should not be held open elsewhere. (Win32)</p>
</li>
<li><a name="umask"></a><b>umask</b>
<p>Returns undef where unavailable, as of version 5.005.</p>
<p><code class="inline"><a class="l_k" href="functions/umask.html">umask</a></code> works but the correct permissions are set only when the file
is finally closed. (AmigaOS)</p>
</li>
<li><a name="utime"></a><b>utime</b>
<p>Only the modification time is updated. (BeOS, Mac OS, VMS, RISC OS)</p>
<p>May not behave as expected.  Behavior depends on the C runtime
library&#39;s implementation of utime(), and the filesystem being
used.  The FAT filesystem typically does not support an &quot;access
time&quot; field, and it may limit timestamps to a granularity of
two seconds. (Win32)</p>
</li>
<li><a name="wait"></a><b>wait</b>
</li>
<li><a name="waitpid"></a><b>waitpid</b>
<p>Not implemented. (Mac OS)</p>
<p>Can only be applied to process handles returned for processes spawned
using <code class="inline"><a class="l_k" href="functions/system.html">system</a><span class="s">(</span><span class="n">1</span><span class="cm">,</span> ...<span class="s">)</span></code>
 or pseudo processes created with <code class="inline"><a class="l_k" href="functions/fork.html">fork()</a></code>. (Win32)</p>
<p>Not useful. (RISC OS)</p>
</li>
</ul>
<a name="Supported-Platforms"></a><h1>Supported Platforms</h1>
<p>As of July 2002 (the Perl release 5.8.0), the following platforms are
able to build Perl from the standard source code distribution
available at <a href="http://www.cpan.org/src/index.html">http://www.cpan.org/src/index.html</a></p>
<pre class="verbatim"><ol><li>        AIX</li><li>        BeOS</li><li>        BSD/OS          (BSDi)</li><li>        Cygwin</li><li>        DG/UX</li><li>        DOS DJGPP       1)</li><li>        DYNIX/ptx</li><li>        EPOC R5</li><li>        FreeBSD</li><li>        HI-UXMPP        (Hitachi) (5.8.0 worked but we didn't know it)</li><li>        HP-UX</li><li>        IRIX</li><li>        Linux</li><li>        Mac OS Classic</li><li>        Mac OS X        (Darwin)</li><li>        MPE/iX</li><li>        NetBSD</li><li>        NetWare</li><li>        NonStop-UX</li><li>        ReliantUNIX     (formerly SINIX)</li><li>        OpenBSD</li><li>        OpenVMS         (formerly VMS)</li><li>        Open UNIX       (Unixware) (since Perl 5.8.1/5.9.0)</li><li>        OS/2</li><li>        OS/400          (using the PASE) (since Perl 5.8.1/5.9.0)</li><li>        PowerUX</li><li>        POSIX-BC        (formerly BS2000)</li><li>        QNX</li><li>        Solaris</li><li>        SunOS 4</li><li>        SUPER-UX        (NEC)</li><li>        Tru64 UNIX      (formerly DEC OSF/1, Digital UNIX)</li><li>        UNICOS</li><li>        UNICOS/mk</li><li>        UTS</li><li>        VOS</li><li>        Win95/98/ME/2K/XP 2)</li><li>        WinCE</li><li>        z/OS            (formerly OS/390)</li><li>        VM/ESA</li><li></li><li>        1) in DOS mode either the DOS or OS/2 ports can be used</li><li>        2) compilers: Borland, MinGW (GCC), VC6</li></ol></pre><p>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 time
for the 5.8.0 release.  There is a very good chance that many of these
will work fine with the 5.8.0.</p>
<pre class="verbatim"><ol><li>        <span class="w">BSD</span>/<span class="w">OS</span></li><li>        <span class="w">DomainOS</span></li><li>        <span class="w">Hurd</span></li><li>        <span class="w">LynxOS</span></li><li>        <span class="w">MachTen</span></li><li>        <span class="w">PowerMAX</span></li><li>        <span class="w">SCO</span> <span class="w">SV</span></li><li>        <span class="w">SVR4</span></li><li>        <span class="w">Unixware</span></li><li>        <span class="w">Windows</span> <span class="n">3.1</span></li></ol></pre><p>Known to be broken for 5.8.0 (but 5.6.1 and 5.7.2 can be used):</p>
<pre class="verbatim"><ol><li>	<span class="w">AmigaOS</span></li></ol></pre><p>The following platforms have been known to build Perl from source in
the past (5.005_03 and earlier), but we haven&#39;t been able to verify
their status for the current release, either because the
hardware/software platforms are rare or because we don&#39;t have an
active champion on these platforms--or both.  They used to work,
though, so go ahead and try compiling them, and let perlbug@perl.org
of any trouble.</p>
<pre class="verbatim"><ol><li>        3b1</li><li>        A/UX</li><li>        ConvexOS</li><li>        CX/UX</li><li>        DC/OSx</li><li>        DDE SMES</li><li>        DOS EMX</li><li>        Dynix</li><li>        EP/IX</li><li>        ESIX</li><li>        FPS</li><li>        GENIX</li><li>        Greenhills</li><li>        ISC</li><li>        MachTen 68k</li><li>        MiNT</li><li>        MPC</li><li>        NEWS-OS</li><li>        NextSTEP</li><li>        OpenSTEP</li><li>        Opus</li><li>        Plan 9</li><li>        RISC/os</li><li>        SCO ODT/OSR</li><li>        Stellar</li><li>        SVR2</li><li>        TI1500</li><li>        TitanOS</li><li>        Ultrix</li><li>        Unisys Dynix</li></ol></pre><p>The following platforms have their own source code distributions and
binaries available via <a href="http://www.cpan.org/ports/">http://www.cpan.org/ports/</a></p>
<pre class="verbatim"><ol><li>                                Perl release</li><li></li><li>        OS/400 (ILE)            5.005_02</li><li>        Tandem Guardian         5.004</li></ol></pre><p>The following platforms have only binaries available via
<a href="http://www.cpan.org/ports/index.html">http://www.cpan.org/ports/index.html</a> :</p>
<pre class="verbatim"><ol><li>                                Perl release</li><li></li><li>        Acorn RISCOS            5.005_02</li><li>        AOS                     5.002</li><li>        LynxOS                  5.004_02</li></ol></pre><p>Although we do suggest that you always build your own Perl from
the source code, both for maximal configurability and for security,
in case you are in a hurry you can check
<a href="http://www.cpan.org/ports/index.html">http://www.cpan.org/ports/index.html</a> for binary distributions.</p>
<a name="SEE-ALSO"></a><h1>SEE ALSO</h1>
<p><a href="perlaix.html">perlaix</a>, <a href="perlamiga.html">perlamiga</a>, <a href="perlapollo.html">perlapollo</a>, <a href="perlbeos.html">perlbeos</a>, <a href="perlbs2000.html">perlbs2000</a>,
<a href="perlce.html">perlce</a>, <a href="perlcygwin.html">perlcygwin</a>, <a href="perldgux.html">perldgux</a>, <a href="perldos.html">perldos</a>, <a href="perlepoc.html">perlepoc</a>,
<a href="perlebcdic.html">perlebcdic</a>, <a href="perlfreebsd.html">perlfreebsd</a>, <a href="perlhurd.html">perlhurd</a>, <a href="perlhpux.html">perlhpux</a>, <a href="perlirix.html">perlirix</a>,
<a href="perlmachten.html">perlmachten</a>, <a href="perlmacos.html">perlmacos</a>, <a href="perlmacosx.html">perlmacosx</a>, <a href="perlmint.html">perlmint</a>, <a href="perlmpeix.html">perlmpeix</a>,
<a href="perlnetware.html">perlnetware</a>, <a href="perlos2.html">perlos2</a>, <a href="perlos390.html">perlos390</a>, <a href="perlos400.html">perlos400</a>,
<a href="perlplan9.html">perlplan9</a>, <a href="perlqnx.html">perlqnx</a>, <a href="perlsolaris.html">perlsolaris</a>, <a href="perltru64.html">perltru64</a>,
<a href="perlunicode.html">perlunicode</a>, <a href="perlvmesa.html">perlvmesa</a>, <a href="perlvms.html">perlvms</a>, <a href="perlvos.html">perlvos</a>,
<a href="perlwin32.html">perlwin32</a>, and <a href="http://search.cpan.org/perldoc/Win32">Win32</a>.</p>
<a name="AUTHORS-%2f-CONTRIBUTORS"></a><h1>AUTHORS / CONTRIBUTORS</h1>
<p>Abigail &lt;abigail@foad.org&gt;,
Charles Bailey &lt;bailey@newman.upenn.edu&gt;,
Graham Barr &lt;gbarr@pobox.com&gt;,
Tom Christiansen &lt;tchrist@perl.com&gt;,
Nicholas Clark &lt;nick@ccl4.org&gt;,
Thomas Dorner &lt;Thomas.Dorner@start.de&gt;,
Andy Dougherty &lt;doughera@lafayette.edu&gt;,
Dominic Dunlop &lt;domo@computer.org&gt;,
Neale Ferguson &lt;neale@vma.tabnsw.com.au&gt;,
David J. Fiander &lt;davidf@mks.com&gt;,
Paul Green &lt;Paul.Green@stratus.com&gt;,
M.J.T. Guy &lt;mjtg@cam.ac.uk&gt;,
Jarkko Hietaniemi &lt;jhi@iki.fi&gt;,
Luther Huffman &lt;lutherh@stratcom.com&gt;,
Nick Ing-Simmons &lt;nick@ing-simmons.net&gt;,
Andreas J. Knig &lt;a.koenig@mind.de&gt;,
Markus Laker &lt;mlaker@contax.co.uk&gt;,
Andrew M. Langmead &lt;aml@world.std.com&gt;,
Larry Moore &lt;ljmoore@freespace.net&gt;,
Paul Moore &lt;Paul.Moore@uk.origin-it.com&gt;,
Chris Nandor &lt;pudge@pobox.com&gt;,
Matthias Neeracher &lt;neeracher@mac.com&gt;,
Philip Newton &lt;pne@cpan.org&gt;,
Gary Ng &lt;71564.1743@CompuServe.COM&gt;,
Tom Phoenix &lt;rootbeer@teleport.com&gt;,
Andr Pirard &lt;A.Pirard@ulg.ac.be&gt;,
Peter Prymmer &lt;pvhp@forte.com&gt;,
Hugo van der Sanden &lt;hv@crypt0.demon.co.uk&gt;,
Gurusamy Sarathy &lt;gsar@activestate.com&gt;,
Paul J. Schinder &lt;schinder@pobox.com&gt;,
Michael G Schwern &lt;schwern@pobox.com&gt;,
Dan Sugalski &lt;dan@sidhe.org&gt;,
Nathan Torkington &lt;gnat@frii.com&gt;.
John Malmberg &lt;wb8tyw@qsl.net&gt;</p>




  <div id="page_index">
    <div id="page_index_header">
      <div id="page_index_close"><a href="#" onClick="pageIndex.hide();return false;"></a></div>
      <div id="page_index_title"><span class="page_index_top">Page index</span></div>
      <div id="page_index_topright"></div>
    </div>
    <div id="page_index_content">
      <ul><li><a href="#NAME">NAME</a><li><a href="#DESCRIPTION">DESCRIPTION</a><li><a href="#ISSUES">ISSUES</a><ul><li><a href="#Newlines">Newlines</a><li><a href="#Numbers-endianness-and-Width">Numbers endianness and Width</a><li><a href="#Files-and-Filesystems">Files and Filesystems</a><li><a href="#System-Interaction">System Interaction</a><li><a href="#Command-names-versus-file-pathnames">Command names versus file pathnames</a><li><a href="#Networking">Networking</a><li><a href="#Interprocess-Communication-(IPC)">Interprocess Communication (IPC)</a><li><a href="#External-Subroutines-(XS)">External Subroutines (XS)</a><li><a href="#Standard-Modules">Standard Modules</a><li><a href="#Time-and-Date">Time and Date</a><li><a href="#Character-sets-and-character-encoding">Character sets and character encoding</a><li><a href="#Internationalisation">Internationalisation</a><li><a href="#System-Resources">System Resources</a><li><a href="#Security">Security</a><li><a href="#Style">Style</a></ul><li><a href="#CPAN-Testers">CPAN Testers</a><li><a href="#PLATFORMS">PLATFORMS</a><ul><li><a href="#Unix">Unix</a><li><a href="#DOS-and-Derivatives">DOS and Derivatives</a><li><a href="#Mac-OS">Mac OS</a><li><a href="#VMS">VMS</a><li><a href="#VOS">VOS</a><li><a href="#EBCDIC-Platforms">EBCDIC Platforms</a><li><a href="#Acorn-RISC-OS">Acorn RISC OS</a><li><a href="#Other-perls">Other perls</a></ul><li><a href="#FUNCTION-IMPLEMENTATIONS">FUNCTION IMPLEMENTATIONS</a><ul><li><a href="#Alphabetical-Listing-of-Perl-Functions">Alphabetical Listing of Perl Functions</a></ul><li><a href="#Supported-Platforms">Supported Platforms</a><li><a href="#SEE-ALSO">SEE ALSO</a><li><a href="#AUTHORS-%2f-CONTRIBUTORS">AUTHORS / CONTRIBUTORS</a></ul>
    </div>
    <div id="page_index_footer">
      <div id="page_index_bottomleft"></div>
      <div id="page_index_bottom"><span class="page_index_bottom"></span></div>
      <div id="page_index_resize"></div>
    </div>
  </div>


	    &nbsp;
          </div>
          <div id="content_footer">
          </div>
        </div>
        <div id="right_column">
          <div class="side_group">
	    
            <div class="side_panel tools_panel">
              <p>Recently read</p>
              <div id="recent_pages"></div>
            </div>
            <div class="side_panel tools_panel">
              <p>Tools</p>
              <ul>
		<li><a href="preferences.html">Preferences</a>
              </ul>
            </div>
          </div>
        </div>
        <div class="clear"></div>
      </div>
      
      <div id="footer">
	<div id="footer_content">
	  
	</div>
      </div>
      
    </div>
      <script language="JavaScript" src="static/combined-20090809.js" type="text/javascript"></script>
<script language="JavaScript" type="text/javascript">
  perldoc.setPath(0);
  perldoc.pageName    = 'perlport';
  perldoc.pageAddress = 'perlport.html';
  perldoc.contentPage = 1;
</script>
    
  </body>
</html>