diff options
Diffstat (limited to 'upstream/mageia-cauldron/man1/perlport.1')
| -rw-r--r-- | upstream/mageia-cauldron/man1/perlport.1 | 2552 |
1 files changed, 2552 insertions, 0 deletions
diff --git a/upstream/mageia-cauldron/man1/perlport.1 b/upstream/mageia-cauldron/man1/perlport.1 new file mode 100644 index 00000000..20012c74 --- /dev/null +++ b/upstream/mageia-cauldron/man1/perlport.1 @@ -0,0 +1,2552 @@ +.\" -*- mode: troff; coding: utf-8 -*- +.\" Automatically generated by Pod::Man 5.01 (Pod::Simple 3.43) +.\" +.\" Standard preamble: +.\" ======================================================================== +.de Sp \" Vertical space (when we can't use .PP) +.if t .sp .5v +.if n .sp +.. +.de Vb \" Begin verbatim text +.ft CW +.nf +.ne \\$1 +.. +.de Ve \" End verbatim text +.ft R +.fi +.. +.\" \*(C` and \*(C' are quotes in nroff, nothing in troff, for use with C<>. +.ie n \{\ +. ds C` "" +. ds C' "" +'br\} +.el\{\ +. ds C` +. ds C' +'br\} +.\" +.\" Escape single quotes in literal strings from groff's Unicode transform. +.ie \n(.g .ds Aq \(aq +.el .ds Aq ' +.\" +.\" If the F register is >0, we'll generate index entries on stderr for +.\" titles (.TH), headers (.SH), subsections (.SS), items (.Ip), and index +.\" entries marked with X<> in POD. Of course, you'll have to process the +.\" output yourself in some meaningful fashion. +.\" +.\" Avoid warning from groff about undefined register 'F'. +.de IX +.. +.nr rF 0 +.if \n(.g .if rF .nr rF 1 +.if (\n(rF:(\n(.g==0)) \{\ +. if \nF \{\ +. de IX +. tm Index:\\$1\t\\n%\t"\\$2" +.. +. if !\nF==2 \{\ +. nr % 0 +. nr F 2 +. \} +. \} +.\} +.rr rF +.\" ======================================================================== +.\" +.IX Title "PERLPORT 1" +.TH PERLPORT 1 2023-11-28 "perl v5.38.2" "Perl Programmers Reference Guide" +.\" For nroff, turn off justification. Always turn off hyphenation; it makes +.\" way too many mistakes in technical documents. +.if n .ad l +.nh +.SH NAME +perlport \- Writing portable Perl +.SH DESCRIPTION +.IX Header "DESCRIPTION" +Perl runs on numerous operating systems. While most of them share +much in common, they also have their own unique features. +.PP +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. +.PP +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 has 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. +.PP +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. +.PP +Be aware of two important points: +.IP "Not all Perl programs have to be portable" 4 +.IX Item "Not all Perl programs have to be portable" +There is no reason you should not use Perl as a language to glue Unix +tools together, or to prototype a Macintosh application, or to manage the +Windows registry. If it makes no sense to aim for portability for one +reason or another in a given program, then don't bother. +.IP "Nearly all of Perl already \fIis\fR portable" 4 +.IX Item "Nearly all of Perl already is portable" +Don't be fooled into thinking that it is hard to create portable Perl +code. It isn't. Perl tries its level-best to bridge the gaps between +what's available on different platforms, and all the means available to +use those features. Thus almost all Perl code runs on any machine +without modification. But there are some significant issues in +writing portable code, and this document is entirely about those issues. +.PP +Here's the general rule: When you approach a task commonly done +using a whole range of platforms, think about writing portable +code. That way, you don't sacrifice much by way of the implementation +choices you can avail yourself of, and at the same time 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, +VMS, etc.), consider writing platform-specific code. +.PP +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. +.PP +The material below is separated into three main sections: main issues of +portability ("ISSUES"), platform-specific issues ("PLATFORMS"), and +built-in Perl functions that behave differently on various ports +("FUNCTION IMPLEMENTATIONS"). +.PP +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 +(\f(CW\*(C`<IMG SRC="yellow_sign.gif" ALT="Under Construction">\*(C'\fR). +.SH ISSUES +.IX Header "ISSUES" +.SS Newlines +.IX Subsection "Newlines" +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 \f(CW\*(C`\e012\*(C'\fR, one type of DOSish I/O uses \f(CW\*(C`\e015\e012\*(C'\fR, +Mac\ OS uses \f(CW\*(C`\e015\*(C'\fR, and z/OS uses \f(CW\*(C`\e025\*(C'\fR. +.PP +Perl uses \f(CW\*(C`\en\*(C'\fR to represent the "logical" newline, where what is +logical may depend on the platform in use. In MacPerl, \f(CW\*(C`\en\*(C'\fR always +means \f(CW\*(C`\e015\*(C'\fR. On EBCDIC platforms, \f(CW\*(C`\en\*(C'\fR could be \f(CW\*(C`\e025\*(C'\fR or \f(CW\*(C`\e045\*(C'\fR. +In DOSish perls, \f(CW\*(C`\en\*(C'\fR usually means \f(CW\*(C`\e012\*(C'\fR, but when +accessing a file in "text" mode, perl uses the \f(CW\*(C`:crlf\*(C'\fR layer that +translates it to (or from) \f(CW\*(C`\e015\e012\*(C'\fR, depending on whether you're +reading or writing. Unix does the same thing on ttys in canonical +mode. \f(CW\*(C`\e015\e012\*(C'\fR is commonly referred to as CRLF. +.PP +To trim trailing newlines from text lines use +\&\f(CW\*(C`chomp\*(C'\fR. With default settings that function +looks for a trailing \f(CW\*(C`\en\*(C'\fR character and thus trims in a portable way. +.PP +When dealing with binary files (or text files in binary mode) be sure +to explicitly set \f(CW$/\fR to the appropriate value for +your file format before using \f(CW\*(C`chomp\*(C'\fR. +.PP +Because of the "text" mode translation, DOSish perls have limitations in +using \f(CW\*(C`seek\*(C'\fR and +\&\f(CW\*(C`tell\*(C'\fR on a file accessed in "text" mode. +Stick to \f(CW\*(C`seek\*(C'\fR\-ing to +locations you got from \f(CW\*(C`tell\*(C'\fR (and no +others), and you are usually free to use +\&\f(CW\*(C`seek\*(C'\fR and +\&\f(CW\*(C`tell\*(C'\fR even in "text" mode. Using +\&\f(CW\*(C`seek\*(C'\fR or +\&\f(CW\*(C`tell\*(C'\fR or other file operations may be +non-portable. If you use \f(CW\*(C`binmode\*(C'\fR on a +file, however, you can usually +\&\f(CW\*(C`seek\*(C'\fR and +\&\f(CW\*(C`tell\*(C'\fR with arbitrary values safely. +.PP +A common misconception in socket programming is that \f(CW\*(C`\en\ eq\ \e012\*(C'\fR +everywhere. When using protocols such as common Internet protocols, +\&\f(CW\*(C`\e012\*(C'\fR and \f(CW\*(C`\e015\*(C'\fR are called for specifically, and the values of +the logical \f(CW\*(C`\en\*(C'\fR and \f(CW\*(C`\er\*(C'\fR (carriage return) are not reliable. +.PP +.Vb 2 +\& print $socket "Hi there, client!\er\en"; # WRONG +\& print $socket "Hi there, client!\e015\e012"; # RIGHT +.Ve +.PP +However, using \f(CW\*(C`\e015\e012\*(C'\fR (or \f(CW\*(C`\ecM\ecJ\*(C'\fR, or \f(CW\*(C`\ex0D\ex0A\*(C'\fR) can be tedious +and unsightly, as well as confusing to those maintaining the code. As +such, the \f(CW\*(C`Socket\*(C'\fR module supplies the Right Thing for those +who want it. +.PP +.Vb 2 +\& use Socket qw(:DEFAULT :crlf); +\& print $socket "Hi there, client!$CRLF" # RIGHT +.Ve +.PP +When reading from a socket, remember that the default input record +separator \f(CW$/\fR is \f(CW\*(C`\en\*(C'\fR, but robust socket code +will recognize as either \f(CW\*(C`\e012\*(C'\fR or \f(CW\*(C`\e015\e012\*(C'\fR as end of line: +.PP +.Vb 3 +\& while (<$socket>) { # NOT ADVISABLE! +\& # ... +\& } +.Ve +.PP +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: +.PP +.Vb 2 +\& use Socket qw(:DEFAULT :crlf); +\& local($/) = LF; # not needed if $/ is already \e012 +\& +\& while (<$socket>) { +\& s/$CR?$LF/\en/; # not sure if socket uses LF or CRLF, OK +\& # s/\e015?\e012/\en/; # same thing +\& } +.Ve +.PP +This example is preferred over the previous one\-\-even for Unix +platforms\-\-because now any \f(CW\*(C`\e015\*(C'\fR's (\f(CW\*(C`\ecM\*(C'\fR's) are stripped out +(and there was much rejoicing). +.PP +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've not yet been translated to the local +newline representation. A single line of code will often suffice: +.PP +.Vb 2 +\& $data =~ s/\e015?\e012/\en/g; +\& return $data; +.Ve +.PP +Some of this may be confusing. Here's a handy reference to the ASCII CR +and LF characters. You can print it out and stick it in your wallet. +.PP +.Vb 2 +\& LF eq \e012 eq \ex0A eq \ecJ eq chr(10) eq ASCII 10 +\& CR eq \e015 eq \ex0D eq \ecM eq chr(13) eq ASCII 13 +\& +\& | Unix | DOS | Mac | +\& \-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\- +\& \en | LF | LF | CR | +\& \er | CR | CR | LF | +\& \en * | LF | CRLF | CR | +\& \er * | CR | CR | LF | +\& \-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\- +\& * text\-mode STDIO +.Ve +.PP +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 +"\en", and "\en" on output becomes CRLF. +.PP +These are just the most common definitions of \f(CW\*(C`\en\*(C'\fR and \f(CW\*(C`\er\*(C'\fR 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 "Unix" but the code numbers change: +.PP +.Vb 4 +\& LF eq \e025 eq \ex15 eq \ecU eq chr(21) eq CP\-1047 21 +\& LF eq \e045 eq \ex25 eq chr(37) eq CP\-0037 37 +\& CR eq \e015 eq \ex0D eq \ecM eq chr(13) eq CP\-1047 13 +\& CR eq \e015 eq \ex0D eq \ecM eq chr(13) eq CP\-0037 13 +\& +\& | z/OS | OS/400 | +\& \-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\- +\& \en | LF | LF | +\& \er | CR | CR | +\& \en * | LF | LF | +\& \er * | CR | CR | +\& \-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\- +\& * text\-mode STDIO +.Ve +.SS "Numbers endianness and Width" +.IX Subsection "Numbers endianness and Width" +Different CPUs store integers and floating point numbers in different +orders (called \fIendianness\fR) 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 "live" via network connection, or by storing the +numbers to secondary storage such as a disk file or tape. +.PP +Conflicting storage orders make an 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 \f(CW\*(C`pack\*(C'\fR and +\&\f(CW\*(C`unpack\*(C'\fR formats \f(CW\*(C`n\*(C'\fR and \f(CW\*(C`N\*(C'\fR, the +"network" orders. These are guaranteed to be portable. +.PP +As of Perl 5.10.0, you can also use the \f(CW\*(C`>\*(C'\fR and \f(CW\*(C`<\*(C'\fR 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. +.PP +You can explore the endianness of your platform by unpacking a +data structure packed in native format such as: +.PP +.Vb 3 +\& print unpack("h*", pack("s2", 1, 2)), "\en"; +\& # \*(Aq10002000\*(Aq on e.g. Intel x86 or Alpha 21064 in little\-endian mode +\& # \*(Aq00100020\*(Aq on e.g. Motorola 68040 +.Ve +.PP +If you need to distinguish between endian architectures you could use +either of the variables set like so: +.PP +.Vb 2 +\& $is_big_endian = unpack("h*", pack("s", 1)) =~ /01/; +\& $is_little_endian = unpack("h*", pack("s", 1)) =~ /^1/; +.Ve +.PP +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. +.PP +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 +\&\f(CW\*(C`Data::Dumper\*(C'\fR and \f(CW\*(C`Storable\*(C'\fR (included as +of Perl 5.8). Keeping all data as text significantly simplifies matters. +.SS "Files and Filesystems" +.IX Subsection "Files and Filesystems" +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 "path" to uniquely identify a file on the system. How +that path is really written, though, differs considerably. +.PP +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. +.PP +DOS, OS/2, VMS, VOS, and Windows can work similarly to Unix with \f(CW\*(C`/\*(C'\fR +as path separator, or in their own idiosyncratic ways (such as having +several root directories and various "unrooted" device files such NIL: +and LPT:). +.PP +Mac\ OS 9 and earlier used \f(CW\*(C`:\*(C'\fR as a path separator instead of \f(CW\*(C`/\*(C'\fR. +.PP +The filesystem may support neither hard links +(\f(CW\*(C`link\*(C'\fR) nor symbolic links +(\f(CW\*(C`symlink\*(C'\fR, +\&\f(CW\*(C`readlink\*(C'\fR, +\&\f(CW\*(C`lstat\*(C'\fR). +.PP +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). +.PP +The "inode change timestamp" (the \f(CW\*(C`\-C\*(C'\fR +filetest) may really be the "creation timestamp" (which it is not in +Unix). +.PP +VOS perl can emulate Unix filenames with \f(CW\*(C`/\*(C'\fR as path separator. The +native pathname characters greater-than, less-than, number-sign, and +percent-sign are always accepted. +.PP +RISC\ OS perl can emulate Unix filenames with \f(CW\*(C`/\*(C'\fR as path +separator, or go native and use \f(CW\*(C`.\*(C'\fR for path separator and \f(CW\*(C`:\*(C'\fR to +signal filesystems and disk names. +.PP +Don'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 \f(CW\*(C`r\*(C'\fR, \f(CW\*(C`w\*(C'\fR, and \f(CW\*(C`x\*(C'\fR mean on +a directory) are the Unix ones. The various Unix/POSIX compatibility +layers usually try to make interfaces like \f(CW\*(C`chmod\*(C'\fR +work, but sometimes there simply is no good mapping. +.PP +The \f(CW\*(C`File::Spec\*(C'\fR modules provide methods to manipulate path +specifications and return the results in native format for each +platform. This is often unnecessary as Unix-style paths are +understood by Perl on every supported platform, but if you need to +produce native paths for a native utility that does not understand +Unix syntax, or if you are operating on paths or path components +in unknown (and thus possibly native) syntax, \f(CW\*(C`File::Spec\*(C'\fR +is your friend. Here are two brief examples: +.PP +.Vb 2 +\& use File::Spec::Functions; +\& chdir(updir()); # go up one directory +\& +\& # Concatenate a path from its components +\& my $file = catfile(updir(), \*(Aqtemp\*(Aq, \*(Aqfile.txt\*(Aq); +\& # on Unix: \*(Aq../temp/file.txt\*(Aq +\& # on Win32: \*(Aq..\etemp\efile.txt\*(Aq +\& # on VMS: \*(Aq[\-.temp]file.txt\*(Aq +.Ve +.PP +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. +.PP +This is especially noticeable in scripts like Makefiles and test suites, +which often assume \f(CW\*(C`/\*(C'\fR as a path separator for subdirectories. +.PP +Also of use is \f(CW\*(C`File::Basename\*(C'\fR from the standard +distribution, which splits a pathname into pieces (base filename, full +path to directory, and file suffix). +.PP +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 \fI/etc/passwd\fR, +\&\fI/etc/sendmail.conf\fR, \fI/etc/resolv.conf\fR, or even \fI/tmp/\fR. For +example, \fI/etc/passwd\fR 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's documentation, then make it easy for +the user to override the default location of the file. +.PP +Don't assume a text file will end with a newline. They should, +but people forget. +.PP +Do not have two files or directories of the same name with different +case, like \fItest.pl\fR and \fITest.pl\fR, as many platforms have +case-insensitive (or at least case-forgiving) filenames. Also, try +not to have non-word characters (except for \f(CW\*(C`.\*(C'\fR) in the names, and +keep them to the 8.3 convention, for maximum portability, onerous a +burden though this may appear. +.PP +Likewise, when using the \f(CW\*(C`AutoSplit\*(C'\fR 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. +.PP +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. +.PP +Many systems (DOS, VMS ODS\-2) cannot have more than one \f(CW\*(C`.\*(C'\fR in their +filenames. +.PP +Don't assume \f(CW\*(C`>\*(C'\fR won't be the first character of a filename. +Always use the three-arg version of +\&\f(CW\*(C`open\*(C'\fR: +.PP +.Vb 1 +\& open my $fh, \*(Aq<\*(Aq, $existing_file) or die $!; +.Ve +.PP +Two-arg \f(CW\*(C`open\*(C'\fR is magic and can +translate characters like \f(CW\*(C`>\*(C'\fR, \f(CW\*(C`<\*(C'\fR, and \f(CW\*(C`|\*(C'\fR in filenames, +which is usually the wrong thing to do. +\&\f(CW\*(C`sysopen\*(C'\fR and three-arg +\&\f(CW\*(C`open\*(C'\fR don't have this problem. +.PP +Don't use \f(CW\*(C`:\*(C'\fR 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 \f(CW\*(C`@\*(C'\fR, \f(CW\*(C`;\*(C'\fR and +\&\f(CW\*(C`|\*(C'\fR. +.PP +Don't assume that in pathnames you can collapse two leading slashes +\&\f(CW\*(C`//\*(C'\fR into one: some networking and clustering filesystems have special +semantics for that. Let the operating system sort it out. +.PP +The \fIportable filename characters\fR as defined by ANSI C are +.PP +.Vb 4 +\& a b c d e f g h i j k l m n o p q r s t u v w x y z +\& A B C D E F G H I J K L M N O P Q R S T U V W X Y Z +\& 0 1 2 3 4 5 6 7 8 9 +\& . _ \- +.Ve +.PP +and \f(CW\*(C`\-\*(C'\fR shouldn'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 \f(CW\*(C`.\*(C'\fR, if any, and to three characters after the +\&\f(CW\*(C`.\*(C'\fR, if any). (And do not use \f(CW\*(C`.\*(C'\fRs in directory names.) +.SS "System Interaction" +.IX Subsection "System Interaction" +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't stay up late worrying about it. +.PP +Some platforms can'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 +\&\f(CW\*(C`close\*(C'\fR files when you are done with them. +Don't \f(CW\*(C`unlink\*(C'\fR or +\&\f(CW\*(C`rename\*(C'\fR an open file. Don't +\&\f(CW\*(C`tie\*(C'\fR or +\&\f(CW\*(C`open\*(C'\fR a file already tied or opened; +\&\f(CW\*(C`untie\*(C'\fR or +\&\f(CW\*(C`close\*(C'\fR it first. +.PP +Don't open the same file more than once at a time for writing, as some +operating systems put mandatory locks on such files. +.PP +Don'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. +.PP +Don't assume that a single \f(CW\*(C`unlink\*(C'\fR completely +gets rid of the file: some filesystems (most notably the ones in VMS) have +versioned filesystems, and \f(CW\*(C`unlink\*(C'\fR removes only +the most recent one (it doesn't remove all the versions because by default +the native tools on those platforms remove just the most recent version, +too). The portable idiom to remove all the versions of a file is +.PP +.Vb 1 +\& 1 while unlink "file"; +.Ve +.PP +This will terminate if the file is undeletable for some reason +(protected, not there, and so on). +.PP +Don't count on a specific environment variable existing in +\&\f(CW%ENV\fR. Don't count on \f(CW%ENV\fR entries +being case-sensitive, or even case-preserving. Don't try to clear +\&\f(CW%ENV\fR by saying \f(CW\*(C`%ENV = ();\*(C'\fR, or, if you really have +to, make it conditional on \f(CW\*(C`$^O ne \*(AqVMS\*(Aq\*(C'\fR since in VMS the +\&\f(CW%ENV\fR table is much more than a per-process key-value +string table. +.PP +On VMS, some entries in the \f(CW%ENV\fR hash are dynamically +created when their key is used on a read if they did not previously +exist. The values for \f(CW$ENV{HOME}\fR, \f(CW$ENV{TERM}\fR, \f(CW$ENV{PATH}\fR, and +\&\f(CW$ENV{USER}\fR, 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 are documented. +.PP +On VMS by default, changes to the \f(CW%ENV\fR hash persist +after perl exits. Subsequent invocations of perl in the same process can +inadvertently inherit environment settings that were meant to be +temporary. +.PP +Don't count on signals or \f(CW%SIG\fR for anything. +.PP +Don't count on filename globbing. Use +\&\f(CW\*(C`opendir\*(C'\fR, +\&\f(CW\*(C`readdir\*(C'\fR, and +\&\f(CW\*(C`closedir\*(C'\fR instead. +.PP +Don't count on per-program environment variables, or per-program current +directories. +.PP +Don't count on specific values of \f(CW$!\fR, neither numeric nor +especially the string values. Users may switch their locales causing +error messages to be translated into their languages. If you can +trust a POSIXish environment, you can portably use the symbols defined +by the \f(CW\*(C`Errno\*(C'\fR module, like \f(CW\*(C`ENOENT\*(C'\fR. And don't trust on the +values of \f(CW$!\fR at all except immediately after a failed +system call. +.SS "Command names versus file pathnames" +.IX Subsection "Command names versus file pathnames" +Don't assume that the name used to invoke a command or program with +\&\f(CW\*(C`system\*(C'\fR or \f(CW\*(C`exec\*(C'\fR 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 "internal" 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, +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 \f(CW\*(C`perl\*(C'\fR might exist in a file named +\&\fIperl\fR, \fIperl.exe\fR, or \fIperl.pm\fR, depending on the operating system. +The variable \f(CW$Config{_exe}\fR in the +\&\f(CW\*(C`Config\*(C'\fR module holds the executable suffix, if any. Third, +the VMS port carefully sets up \f(CW$^X\fR and +\&\f(CW$Config{perlpath}\fR so that no further processing +is required. This is just as well, because the matching regular +expression used below would then have to deal with a possible trailing +version number in the VMS file name. +.PP +To convert \f(CW$^X\fR to a file pathname, taking account of +the requirements of the various operating system possibilities, say: +.PP +.Vb 6 +\& use Config; +\& my $thisperl = $^X; +\& if ($^O ne \*(AqVMS\*(Aq) { +\& $thisperl .= $Config{_exe} +\& unless $thisperl =~ m/\eQ$Config{_exe}\eE$/i; +\& } +.Ve +.PP +To convert \f(CW$Config{perlpath}\fR to a file pathname, say: +.PP +.Vb 6 +\& use Config; +\& my $thisperl = $Config{perlpath}; +\& if ($^O ne \*(AqVMS\*(Aq) { +\& $thisperl .= $Config{_exe} +\& unless $thisperl =~ m/\eQ$Config{_exe}\eE$/i; +\& } +.Ve +.SS Networking +.IX Subsection "Networking" +Don't assume that you can reach the public Internet. +.PP +Don't assume that there is only one way to get through firewalls +to the public Internet. +.PP +Don't assume that you can reach outside world through any other port +than 80, or some web proxy. ftp is blocked by many firewalls. +.PP +Don't assume that you can send email by connecting to the local SMTP port. +.PP +Don't assume that you can reach yourself or any node by the name +\&'localhost'. The same goes for '127.0.0.1'. You will have to try both. +.PP +Don't assume that the host has only one network card, or that it +can't bind to many virtual IP addresses. +.PP +Don't assume a particular network device name. +.PP +Don't assume a particular set of +\&\f(CW\*(C`ioctl\*(C'\fRs will work. +.PP +Don't assume that you can ping hosts and get replies. +.PP +Don't assume that any particular port (service) will respond. +.PP +Don't assume that \f(CW\*(C`Sys::Hostname\*(C'\fR (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 that for things such as DHCP and NAT, the hostname you get back +might not be very useful. +.PP +All the above \fIdon't\fRs 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. +.SS "Interprocess Communication (IPC)" +.IX Subsection "Interprocess Communication (IPC)" +In general, don't directly access the system in code meant to be +portable. That means, no \f(CW\*(C`system\*(C'\fR, +\&\f(CW\*(C`exec\*(C'\fR, \f(CW\*(C`fork\*(C'\fR, +\&\f(CW\*(C`pipe\*(C'\fR, +\&\f(CW\`\`\fR or \f(CW\*(C`qx//\*(C'\fR, +\&\f(CW\*(C`open\*(C'\fR with a \f(CW\*(C`|\*(C'\fR, nor any of the other +things that makes being a Perl hacker worth being. +.PP +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're calling +\&\f(CW\*(C`netstat \-a\*(C'\fR, you probably don't expect it to run on both Unix and CP/M.) +.PP +One especially common bit of Perl code is opening a pipe to \fBsendmail\fR: +.PP +.Vb 2 +\& open(my $mail, \*(Aq|\-\*(Aq, \*(Aq/usr/lib/sendmail \-t\*(Aq) +\& or die "cannot fork sendmail: $!"; +.Ve +.PP +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. \f(CW\*(C`Mail::Mailer\*(C'\fR and \f(CW\*(C`Mail::Send\*(C'\fR +in the \f(CW\*(C`MailTools\*(C'\fR distribution are commonly used, and provide several +mailing methods, including \f(CW\*(C`mail\*(C'\fR, \f(CW\*(C`sendmail\*(C'\fR, and direct SMTP (via +\&\f(CW\*(C`Net::SMTP\*(C'\fR) if a mail transfer agent is not available. +\&\f(CW\*(C`Mail::Sendmail\*(C'\fR is a standalone module that provides +simple, platform-independent mailing. +.PP +The Unix System V IPC (\f(CW\*(C`msg*(), sem*(), shm*()\*(C'\fR) is not available +even on all Unix platforms. +.PP +Do not use either the bare result of \f(CW\*(C`pack("N", 10, 20, 30, 40)\*(C'\fR or +bare v\-strings (such as \f(CW\*(C`v10.20.30.40\*(C'\fR) to represent IPv4 addresses: +both forms just pack the four bytes into network order. That this +would be equal to the C language \f(CW\*(C`in_addr\*(C'\fR struct (which is what the +socket code internally uses) is not guaranteed. To be portable use +the routines of the \f(CW\*(C`Socket\*(C'\fR module, such as +\&\f(CW\*(C`inet_aton\*(C'\fR, +\&\f(CW\*(C`inet_ntoa\*(C'\fR, and +\&\f(CW\*(C`sockaddr_in\*(C'\fR. +.PP +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 exposes a common interface). +.SS "External Subroutines (XS)" +.IX Subsection "External Subroutines (XS)" +XS code can usually be made to work with any platform, but dependent +libraries, header files, etc., might not be readily available or +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. +.PP +A different type of portability issue arises when writing XS code: +availability of a C compiler on the end-user's system. C brings +with it its own portability issues, and writing XS code will expose +you to some of those. Writing purely in Perl is an easier way to +achieve portability. +.SS "Standard Modules" +.IX Subsection "Standard Modules" +In general, the standard modules work across platforms. Notable +exceptions are the \f(CW\*(C`CPAN\*(C'\fR module (which currently makes +connections to external programs that may not be available), +platform-specific modules (like \f(CW\*(C`ExtUtils::MM_VMS\*(C'\fR), +and DBM modules. +.PP +There is no one DBM module available on all platforms. +\&\f(CW\*(C`SDBM_File\*(C'\fR and the others are generally available on all +Unix and DOSish ports, but not in MacPerl, where only +\&\f(CW\*(C`NDBM_File\*(C'\fR and \f(CW\*(C`DB_File\*(C'\fR are available. +.PP +The good news is that at least some DBM module should be available, and +\&\f(CW\*(C`AnyDBM_File\*(C'\fR 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 AnyDBM_File for more details. +.SS "Time and Date" +.IX Subsection "Time and Date" +The system's notion of time of day and calendar date is controlled in +widely different ways. Don't assume the timezone is stored in \f(CW$ENV{TZ}\fR, +and even if it is, don't assume that you can control the timezone through +that variable. Don't assume anything about the three-letter timezone +abbreviations (for example that MST would be the Mountain Standard Time, +it'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. +.PP +Don't assume that the epoch starts at 00:00:00, January 1, 1970, +because that is OS\- and implementation-specific. It is better to +store a date in an unambiguous representation. The ISO 8601 standard +defines YYYY-MM-DD as the date format, or YYYY\-MM\-DDTHH:MM:SS +(that's a literal "T" separating the date from the time). +Please do use the ISO 8601 instead of making us guess what +date 02/03/04 might be. ISO 8601 even sorts nicely as-is. +A text representation (like "1987\-12\-18") can be easily converted +into an OS-specific value using a module like +\&\f(CW\*(C`Time::Piece\*(C'\fR (see "Date Parsing" in Time::Piece) or +\&\f(CW\*(C`Date::Parse\*(C'\fR. An array of values, such as those +returned by \f(CW\*(C`localtime\*(C'\fR, can be converted to an OS-specific +representation using \f(CW\*(C`Time::Local\*(C'\fR. +.PP +When calculating specific times, such as for tests in time or date modules, +it may be appropriate to calculate an offset for the epoch. +.PP +.Vb 2 +\& use Time::Local qw(timegm); +\& my $offset = timegm(0, 0, 0, 1, 0, 1970); +.Ve +.PP +The value for \f(CW$offset\fR in Unix will be \f(CW0\fR, but in Mac OS Classic +will be some large number. \f(CW$offset\fR can then be added to a Unix time +value to get what should be the proper value on any system. +.SS "Character sets and character encoding" +.IX Subsection "Character sets and character encoding" +Assume very little about character sets. +.PP +Assume nothing about numerical values (\f(CW\*(C`ord\*(C'\fR, +\&\f(CW\*(C`chr\*(C'\fR) of characters. +Do not use explicit code point ranges (like \f(CW\*(C`\exHH\-\exHH)\*(C'\fR. However, +starting in Perl v5.22, regular expression pattern bracketed character +class ranges specified like \f(CW\*(C`qr/[\eN{U+HH}\-\eN{U+HH}]/\*(C'\fR are portable, +and starting in Perl v5.24, the same ranges are portable in +\&\f(CW\*(C`tr///\*(C'\fR. +You can portably use symbolic character classes like \f(CW\*(C`[:print:]\*(C'\fR. +.PP +Do not assume that the alphabetic characters are encoded contiguously +(in the numeric sense). There may be gaps. Special coding in Perl, +however, guarantees that all subsets of \f(CW\*(C`qr/[A\-Z]/\*(C'\fR, \f(CW\*(C`qr/[a\-z]/\*(C'\fR, and +\&\f(CW\*(C`qr/[0\-9]/\*(C'\fR behave as expected. +\&\f(CW\*(C`tr///\*(C'\fR +behaves the same for these ranges. In patterns, any ranges specified with +end points using the \f(CW\*(C`\eN{...}\*(C'\fR notations ensures character set +portability, but it is a bug in Perl v5.22 that this isn't true of +\&\f(CW\*(C`tr///\*(C'\fR, +fixed in v5.24. +.PP +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 "a" and "A" +come before "b"; the accented and other international characters may +be interlaced so that ä comes before "b". +Unicode::Collate can be used to sort this all out. +.SS Internationalisation +.IX Subsection "Internationalisation" +If you may assume POSIX (a rather large assumption), you may read +more about the POSIX locale system from perllocale. 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. +.PP +If you really want to be international, you should consider Unicode. +See perluniintro and perlunicode for more information. +.PP +By default Perl assumes your source code is written in an 8\-bit ASCII +superset. To embed Unicode characters in your strings and regexes, you can +use the \f(CW\*(C`\ex{HH}\*(C'\fR or (more portably) \f(CW\*(C`\eN{U+HH}\*(C'\fR +notations. You can also use the +\&\f(CW\*(C`utf8\*(C'\fR pragma and write your code in UTF\-8, which lets you use +Unicode characters directly (not just in quoted constructs but also in +identifiers). +.SS "System Resources" +.IX Subsection "System Resources" +If your code is destined for systems with severely constrained (or +missing!) virtual memory systems then you want to be \fIespecially\fR mindful +of avoiding wasteful constructs such as: +.PP +.Vb 1 +\& my @lines = <$very_large_file>; # bad +\& +\& while (<$fh>) {$file .= $_} # sometimes bad +\& my $file = join(\*(Aq\*(Aq, <$fh>); # better +.Ve +.PP +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 than the first. +.SS Security +.IX Subsection "Security" +Most multi-user platforms provide basic levels of security, usually +implemented at the filesystem level. Some, however, unfortunately do +not. Thus the notion of user id, or "home" directory, +or even the state of being logged-in, may be unrecognizable on many +platforms. If you write 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). +.PP +Don'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 \f(CW\*(C`rwx\*(C'\fR. Even if the \f(CW\*(C`rwx\*(C'\fR exist, +their semantics might be different. +.PP +(From the 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.) +.PP +Don't assume the Unix user and group semantics: especially, don't +expect \f(CW$<\fR and \f(CW$>\fR (or +\&\f(CW$(\fR and \f(CW$)\fR) to work for switching +identities (or memberships). +.PP +Don'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.) +.SS Style +.IX Subsection "Style" +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 \f(CW\*(C`Config\*(C'\fR module and the +special variable \f(CW$^O\fR to differentiate platforms, as +described in "PLATFORMS". +.PP +Beware of the "else syndrome": +.PP +.Vb 5 +\& if ($^O eq \*(AqMSWin32\*(Aq) { +\& # code that assumes Windows +\& } else { +\& # code that assumes Linux +\& } +.Ve +.PP +The \f(CW\*(C`else\*(C'\fR branch should be used for the really ultimate fallback, +not for code specific to some platform. +.PP +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 +\&\f(CW$!\fR after a failed system call. Using +\&\f(CW$!\fR for anything else than displaying it as output is +doubtful (though see the \f(CW\*(C`Errno\*(C'\fR 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't anchor a regex when testing an error value. +.SH "CPAN Testers" +.IX Header "CPAN Testers" +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. +.PP +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. +.PP +Also see: +.IP \(bu 4 +Mailing list: cpan\-testers\-discuss@perl.org +.IP \(bu 4 +Testing results: <https://www.cpantesters.org/> +.SH PLATFORMS +.IX Header "PLATFORMS" +Perl is built with a \f(CW$^O\fR variable that indicates the +operating system it was built on. This was implemented +to help speed up code that would otherwise have to \f(CW\*(C`use Config\*(C'\fR +and use the value of \f(CW$Config{osname}\fR. Of course, +to get more detailed information about the system, looking into +\&\f(CW%Config\fR is certainly recommended. +.PP +\&\f(CW%Config\fR 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. +.SS Unix +.IX Subsection "Unix" +Perl works on a bewildering variety of Unix and Unix-like platforms (see +e.g. most of the files in the \fIhints/\fR directory in the source code kit). +On most of these systems, the value of \f(CW$^O\fR (hence +\&\f(CW$Config{osname}\fR, too) is determined either by +lowercasing and stripping punctuation from the first field of the string +returned by typing \f(CW\*(C`uname \-a\*(C'\fR (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: +.PP +.Vb 10 +\& uname $^O $Config{archname} +\& \-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\- +\& AIX aix aix +\& BSD/OS bsdos i386\-bsdos +\& Darwin darwin darwin +\& DYNIX/ptx dynixptx i386\-dynixptx +\& FreeBSD freebsd freebsd\-i386 +\& Haiku haiku BePC\-haiku +\& Linux linux arm\-linux +\& Linux linux armv5tel\-linux +\& Linux linux i386\-linux +\& Linux linux i586\-linux +\& Linux linux ppc\-linux +\& HP\-UX hpux PA\-RISC1.1 +\& IRIX irix irix +\& Mac OS X darwin darwin +\& NeXT 3 next next\-fat +\& NeXT 4 next OPENSTEP\-Mach +\& openbsd openbsd i386\-openbsd +\& OSF1 dec_osf alpha\-dec_osf +\& reliantunix\-n svr4 RM400\-svr4 +\& SCO_SV sco_sv i386\-sco_sv +\& SINIX\-N svr4 RM400\-svr4 +\& sn4609 unicos CRAY_C90\-unicos +\& sn6521 unicosmk t3e\-unicosmk +\& sn9617 unicos CRAY_J90\-unicos +\& SunOS solaris sun4\-solaris +\& SunOS solaris i86pc\-solaris +\& SunOS4 sunos sun4\-sunos +.Ve +.PP +Because the value of \f(CW$Config{archname}\fR may +depend on the hardware architecture, it can vary more than the value of +\&\f(CW$^O\fR. +.SS "DOS and Derivatives" +.IX Subsection "DOS and Derivatives" +Perl has long been ported to Intel-style microcomputers running under +systems like PC-DOS, MS-DOS, OS/2, and most Windows platforms you can +bring yourself to mention (except for Windows CE, if you count that). +Users familiar with \fICOMMAND.COM\fR or \fICMD.EXE\fR style shells should +be aware that each of these file specifications may have subtle +differences: +.PP +.Vb 4 +\& my $filespec0 = "c:/foo/bar/file.txt"; +\& my $filespec1 = "c:\e\efoo\e\ebar\e\efile.txt"; +\& my $filespec2 = \*(Aqc:\efoo\ebar\efile.txt\*(Aq; +\& my $filespec3 = \*(Aqc:\e\efoo\e\ebar\e\efile.txt\*(Aq; +.Ve +.PP +System calls accept either \f(CW\*(C`/\*(C'\fR or \f(CW\*(C`\e\*(C'\fR as the path separator. +However, many command-line utilities of DOS vintage treat \f(CW\*(C`/\*(C'\fR as +the option prefix, so may get confused by filenames containing \f(CW\*(C`/\*(C'\fR. +Aside from calling any external programs, \f(CW\*(C`/\*(C'\fR will work just fine, +and probably better, as it is more consistent with popular usage, +and avoids the problem of remembering what to backwhack and what +not to. +.PP +The DOS FAT filesystem can accommodate only "8.3" style filenames. Under +the "case-insensitive, but case-preserving" HPFS (OS/2) and NTFS (NT) +filesystems you may have to be careful about case returned with functions +like \f(CW\*(C`readdir\*(C'\fR or used with functions like +\&\f(CW\*(C`open\*(C'\fR or +\&\f(CW\*(C`opendir\*(C'\fR. +.PP +DOS also treats several filenames as special, such as \fIAUX\fR, \fIPRN\fR, +\&\fINUL\fR, \fICON\fR, \fICOM1\fR, \fILPT1\fR, \fILPT2\fR, etc. Unfortunately, sometimes +these filenames won't even work if you include an explicit directory +prefix. It is best to avoid such filenames, if you want your code to be +portable to DOS and its derivatives. It's hard to know what these all +are, unfortunately. +.PP +Users of these operating systems may also wish to make use of +scripts such as \fIpl2bat.bat\fR to put wrappers around your scripts. +.PP +Newline (\f(CW\*(C`\en\*(C'\fR) is translated as \f(CW\*(C`\e015\e012\*(C'\fR by the I/O system when +reading from and writing to files (see "Newlines"). +\&\f(CWbinmode($filehandle)\fR will keep \f(CW\*(C`\en\*(C'\fR translated as \f(CW\*(C`\e012\*(C'\fR for that +filehandle. +\&\f(CW\*(C`binmode\*(C'\fR should always be used for code +that deals with binary data. That's assuming you realize in advance that +your data is in binary. General-purpose programs should often assume +nothing about their data. +.PP +The \f(CW$^O\fR variable and the +\&\f(CW$Config{archname}\fR values for various DOSish +perls are as follows: +.PP +.Vb 10 +\& OS $^O $Config{archname} ID Version +\& \-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\- +\& MS\-DOS dos ? +\& PC\-DOS dos ? +\& OS/2 os2 ? +\& Windows 3.1 ? ? 0 3 01 +\& Windows 95 MSWin32 MSWin32\-x86 1 4 00 +\& Windows 98 MSWin32 MSWin32\-x86 1 4 10 +\& Windows ME MSWin32 MSWin32\-x86 1 ? +\& Windows NT MSWin32 MSWin32\-x86 2 4 xx +\& Windows NT MSWin32 MSWin32\-ALPHA 2 4 xx +\& Windows NT MSWin32 MSWin32\-ppc 2 4 xx +\& Windows 2000 MSWin32 MSWin32\-x86 2 5 00 +\& Windows XP MSWin32 MSWin32\-x86 2 5 01 +\& Windows 2003 MSWin32 MSWin32\-x86 2 5 02 +\& Windows Vista MSWin32 MSWin32\-x86 2 6 00 +\& Windows 7 MSWin32 MSWin32\-x86 2 6 01 +\& Windows 7 MSWin32 MSWin32\-x64 2 6 01 +\& Windows 2008 MSWin32 MSWin32\-x86 2 6 01 +\& Windows 2008 MSWin32 MSWin32\-x64 2 6 01 +\& Windows CE MSWin32 ? 3 +\& Cygwin cygwin cygwin +.Ve +.PP +The various MSWin32 Perl's can distinguish the OS they are running on +via the value of the fifth element of the list returned from +\&\f(CWWin32::GetOSVersion()\fR. For example: +.PP +.Vb 4 +\& if ($^O eq \*(AqMSWin32\*(Aq) { +\& my @os_version_info = Win32::GetOSVersion(); +\& print +(\*(Aq3.1\*(Aq,\*(Aq95\*(Aq,\*(AqNT\*(Aq)[$os_version_info[4]],"\en"; +\& } +.Ve +.PP +There are also \f(CW\*(C`Win32::IsWinNT()|Win32/Win32::IsWinNT()\*(C'\fR, +\&\f(CW\*(C`Win32::IsWin95()|Win32/Win32::IsWin95()\*(C'\fR, and +\&\f(CWWin32::GetOSName()\fR; try +\&\f(CW\*(C`perldoc Win32\*(C'\fR. +The very portable \f(CWPOSIX::uname()\fR will work too: +.PP +.Vb 2 +\& c:\e> perl \-MPOSIX \-we "print join \*(Aq|\*(Aq, uname" +\& Windows NT|moonru|5.0|Build 2195 (Service Pack 2)|x86 +.Ve +.PP +Errors set by Winsock functions are now put directly into \f(CW$^E\fR, +and the relevant \f(CW\*(C`WSAE*\*(C'\fR error codes are now exported from the +Errno and POSIX modules for testing this against. +.PP +The previous behavior of putting the errors (converted to POSIX-style +\&\f(CW\*(C`E*\*(C'\fR error codes since Perl 5.20.0) into \f(CW$!\fR was buggy due to +the non-equivalence of like-named Winsock and POSIX error constants, +a relationship between which has unfortunately been established +in one way or another since Perl 5.8.0. +.PP +The new behavior provides a much more robust solution for checking +Winsock errors in portable software without accidentally matching +POSIX tests that were intended for other OSes and may have different +meanings for Winsock. +.PP +The old behavior is currently retained, warts and all, for backwards +compatibility, but users are encouraged to change any code that +tests \f(CW$!\fR against \f(CW\*(C`E*\*(C'\fR constants for Winsock errors to instead +test \f(CW$^E\fR against \f(CW\*(C`WSAE*\*(C'\fR constants. After a suitable deprecation +period, which started with Perl 5.24, the old behavior may be +removed, leaving \f(CW$!\fR unchanged after Winsock function calls, to +avoid any possible confusion over which error variable to check. +.PP +Also see: +.IP \(bu 4 +The EMX environment for DOS, OS/2, etc. emx@iaehv.nl, +<ftp://hobbes.nmsu.edu/pub/os2/dev/emx/> Also perlos2. +.IP \(bu 4 +Build instructions for Win32 in perlwin32, or under the Cygnus environment +in perlcygwin. +.IP \(bu 4 +The \f(CW\*(C`Win32::*\*(C'\fR modules in Win32. +.IP \(bu 4 +The ActiveState Pages, <https://www.activestate.com/> +.IP \(bu 4 +The Cygwin environment for Win32; \fIREADME.cygwin\fR (installed +as perlcygwin), <https://www.cygwin.com/> +.IP \(bu 4 +Build instructions for OS/2, perlos2 +.SS VMS +.IX Subsection "VMS" +Perl on VMS is discussed in perlvms in the Perl distribution. +.PP +The official name of VMS as of this writing is OpenVMS. +.PP +Interacting with Perl from the Digital Command Language (DCL) shell +often requires a different set of quotation marks than Unix shells do. +For example: +.PP +.Vb 2 +\& $ perl \-e "print ""Hello, world.\en""" +\& Hello, world. +.Ve +.PP +There are several ways to wrap your Perl scripts in DCL \fI.COM\fR files, if +you are so inclined. For example: +.PP +.Vb 6 +\& $ write sys$output "Hello from DCL!" +\& $ if p1 .eqs. "" +\& $ then perl \-x \*(Aqf$environment("PROCEDURE") +\& $ else perl \-x \- \*(Aqp1 \*(Aqp2 \*(Aqp3 \*(Aqp4 \*(Aqp5 \*(Aqp6 \*(Aqp7 \*(Aqp8 +\& $ deck/dollars="_\|_END_\|_" +\& #!/usr/bin/perl +\& +\& print "Hello from Perl!\en"; +\& +\& _\|_END_\|_ +\& $ endif +.Ve +.PP +Do take care with \f(CW\*(C`$ ASSIGN/nolog/user SYS$COMMAND: SYS$INPUT\*(C'\fR if your +Perl-in-DCL script expects to do things like \f(CW\*(C`$read = <STDIN>;\*(C'\fR. +.PP +The VMS operating system has two filesystems, designated by their +on-disk structure (ODS) level: ODS\-2 and its successor ODS\-5. The +initial port of Perl to VMS pre-dates ODS\-5, but all current testing and +development assumes ODS\-5 and its capabilities, including case +preservation, extended characters in filespecs, and names up to 8192 +bytes long. +.PP +Perl on VMS can accept either VMS\- or Unix-style file +specifications as in either of the following: +.PP +.Vb 2 +\& $ perl \-ne "print if /perl_setup/i" SYS$LOGIN:LOGIN.COM +\& $ perl \-ne "print if /perl_setup/i" /sys$login/login.com +.Ve +.PP +but not a mixture of both as in: +.PP +.Vb 2 +\& $ perl \-ne "print if /perl_setup/i" sys$login:/login.com +\& Can\*(Aqt open sys$login:/login.com: file specification syntax error +.Ve +.PP +In general, the easiest path to portability is always to specify +filenames in Unix format unless they will need to be processed by native +commands or utilities. Because of this latter consideration, the +File::Spec module by default returns native format specifications +regardless of input format. This default may be reversed so that +filenames are always reported in Unix format by specifying the +\&\f(CW\*(C`DECC$FILENAME_UNIX_REPORT\*(C'\fR feature logical in the environment. +.PP +The file type, or extension, is always present in a VMS-format file +specification even if it's zero-length. This means that, by default, +\&\f(CW\*(C`readdir\*(C'\fR will return a trailing dot on a +file with no extension, so where you would see \f(CW"a"\fR on Unix you'll see +\&\f(CW"a."\fR on VMS. However, the trailing dot may be suppressed by enabling +the \f(CW\*(C`DECC$READDIR_DROPDOTNOTYPE\*(C'\fR feature in the environment (see the CRTL +documentation on feature logical names). +.PP +What \f(CW\*(C`\en\*(C'\fR represents depends on the type of file opened. It usually +represents \f(CW\*(C`\e012\*(C'\fR but it could also be \f(CW\*(C`\e015\*(C'\fR, \f(CW\*(C`\e012\*(C'\fR, \f(CW\*(C`\e015\e012\*(C'\fR, +\&\f(CW\*(C`\e000\*(C'\fR, \f(CW\*(C`\e040\*(C'\fR, or nothing depending on the file organization and +record format. The \f(CW\*(C`VMS::Stdio\*(C'\fR module provides access to +the special \f(CWfopen()\fR requirements of files with unusual attributes on +VMS. +.PP +The value of \f(CW$^O\fR on OpenVMS is "VMS". To determine the +architecture that you are running on refer to +\&\f(CW$Config{archname}\fR. +.PP +On VMS, perl determines the UTC offset from the \f(CW\*(C`SYS$TIMEZONE_DIFFERENTIAL\*(C'\fR +logical name. Although the VMS epoch began at 17\-NOV\-1858 00:00:00.00, +calls to \f(CW\*(C`localtime\*(C'\fR are adjusted to count +offsets from 01\-JAN\-1970 00:00:00.00, just like Unix. +.PP +Also see: +.IP \(bu 4 +\&\fIREADME.vms\fR (installed as \fIREADME_vms\fR), perlvms +.IP \(bu 4 +vmsperl list, vmsperl\-subscribe@perl.org +.IP \(bu 4 +vmsperl on the web, <http://www.sidhe.org/vmsperl/index.html> +.IP \(bu 4 +VMS Software Inc. web site, <http://www.vmssoftware.com> +.SS VOS +.IX Subsection "VOS" +Perl on VOS (also known as OpenVOS) is discussed in \fIREADME.vos\fR +in the Perl distribution (installed as perlvos). Perl on VOS +can accept either VOS\- or Unix-style file specifications as in +either of the following: +.PP +.Vb 2 +\& $ perl \-ne "print if /perl_setup/i" >system>notices +\& $ perl \-ne "print if /perl_setup/i" /system/notices +.Ve +.PP +or even a mixture of both as in: +.PP +.Vb 1 +\& $ perl \-ne "print if /perl_setup/i" >system/notices +.Ve +.PP +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. +.PP +Older releases of VOS (prior to OpenVOS Release 17.0) limit file +names to 32 or fewer characters, prohibit file names from +starting with a \f(CW\*(C`\-\*(C'\fR character, and prohibit file names from +containing \f(CW\*(C` \*(C'\fR (space) or any character from the set \f(CW\*(C`!#%&\*(Aq()*;<=>?\*(C'\fR. +.PP +Newer releases of VOS (OpenVOS Release 17.0 or later) support a +feature known as extended names. On these releases, file names +can contain up to 255 characters, are prohibited from starting +with a \f(CW\*(C`\-\*(C'\fR character, and the set of prohibited characters is +reduced to \f(CW\*(C`#%*<>?\*(C'\fR. There are +restrictions involving spaces and apostrophes: these characters +must not begin or end a name, nor can they immediately precede or +follow a period. Additionally, a space must not immediately +precede another space or hyphen. Specifically, the following +character combinations are prohibited: space-space, +space-hyphen, period-space, space-period, period-apostrophe, +apostrophe-period, leading or trailing space, and leading or +trailing apostrophe. Although an extended file name is limited +to 255 characters, a path name is still limited to 256 +characters. +.PP +The value of \f(CW$^O\fR on VOS is "vos". To determine the +architecture that you are running on refer to +\&\f(CW$Config{archname}\fR. +.PP +Also see: +.IP \(bu 4 +\&\fIREADME.vos\fR (installed as perlvos) +.IP \(bu 4 +The VOS mailing list. +.Sp +There is no specific mailing list for Perl on VOS. You can contact +the Stratus Technologies Customer Assistance Center (CAC) for your +region, or you can use the contact information located in the +distribution files on the Stratus Anonymous FTP site. +.IP \(bu 4 +Stratus Technologies on the web at <http://www.stratus.com> +.IP \(bu 4 +VOS Open-Source Software on the web at <http://ftp.stratus.com/pub/vos/vos.html> +.SS "EBCDIC Platforms" +.IX Subsection "EBCDIC Platforms" +v5.22 core Perl runs on z/OS (formerly OS/390). Theoretically it could +run on the successors of OS/400 on AS/400 minicomputers as well as +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). +.PP +The rest of this section may need updating, but we don't know what it +should say. Please submit comments to +<https://github.com/Perl/perl5/issues>. +.PP +On the mainframe Perl currently works under the "Unix system +services for OS/390" (formerly known as OpenEdition), VM/ESA OpenEdition, or +the BS200 POSIX-BC system (BS2000 is supported in Perl 5.6 and greater). +See perlos390 for details. Note that for OS/400 there is also a port of +Perl 5.8.1/5.10.0 or later to the PASE which is ASCII-based (as opposed to +ILE which is EBCDIC-based), see perlos400. +.PP +As of R2.5 of USS for OS/390 and Version 2.3 of VM/ESA these Unix +sub-systems do not support the \f(CW\*(C`#!\*(C'\fR 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: +.PP +.Vb 4 +\& : # use perl +\& eval \*(Aqexec /usr/local/bin/perl \-S $0 ${1+"$@"}\*(Aq +\& if 0; +\& #!/usr/local/bin/perl # just a comment really +\& +\& print "Hello from perl!\en"; +.Ve +.PP +OS/390 will support the \f(CW\*(C`#!\*(C'\fR shebang trick in release 2.8 and beyond. +Calls to \f(CW\*(C`system\*(C'\fR and backticks can use POSIX +shell syntax on all S/390 systems. +.PP +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: +.PP +.Vb 3 +\& BEGIN +\& CALL PGM(PERL5/PERL) PARM(\*(Aq/QOpenSys/hello.pl\*(Aq) +\& ENDPGM +.Ve +.PP +This will invoke the Perl script \fIhello.pl\fR in the root of the +QOpenSys file system. On the AS/400 calls to +\&\f(CW\*(C`system\*(C'\fR or backticks must use CL syntax. +.PP +On these platforms, bear in mind that the EBCDIC character set may have +an effect on what happens with some Perl functions (such as +\&\f(CW\*(C`chr\*(C'\fR, \f(CW\*(C`pack\*(C'\fR, +\&\f(CW\*(C`print\*(C'\fR, +\&\f(CW\*(C`printf\*(C'\fR, +\&\f(CW\*(C`ord\*(C'\fR, \f(CW\*(C`sort\*(C'\fR, +\&\f(CW\*(C`sprintf\*(C'\fR, +\&\f(CW\*(C`unpack\*(C'\fR), as +well as bit-fiddling with ASCII constants using operators like +\&\f(CW\*(C`^\*(C'\fR, \f(CW\*(C`&\*(C'\fR and \f(CW\*(C`|\*(C'\fR, not to mention +dealing with socket interfaces to ASCII computers (see "Newlines"). +.PP +Fortunately, most web servers for the mainframe will correctly +translate the \f(CW\*(C`\en\*(C'\fR in the following statement to its ASCII equivalent +(\f(CW\*(C`\er\*(C'\fR is the same under both Unix and z/OS): +.PP +.Vb 1 +\& print "Content\-type: text/html\er\en\er\en"; +.Ve +.PP +The values of \f(CW$^O\fR on some of these platforms include: +.PP +.Vb 5 +\& uname $^O $Config{archname} +\& \-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\- +\& OS/390 os390 os390 +\& OS400 os400 os400 +\& POSIX\-BC posix\-bc BS2000\-posix\-bc +.Ve +.PP +Some simple tricks for determining if you are running on an EBCDIC +platform could include any of the following (perhaps all): +.PP +.Vb 1 +\& if ("\et" eq "\e005") { print "EBCDIC may be spoken here!\en"; } +\& +\& if (ord(\*(AqA\*(Aq) == 193) { print "EBCDIC may be spoken here!\en"; } +\& +\& if (chr(169) eq \*(Aqz\*(Aq) { print "EBCDIC may be spoken here!\en"; } +.Ve +.PP +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). +.PP +Also see: +.IP \(bu 4 +perlos390, perlos400, perlbs2000, perlebcdic. +.IP \(bu 4 +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 +"subscribe perl-mvs" to majordomo@perl.org. +.IP \(bu 4 +AS/400 Perl information at +<http://as400.rochester.ibm.com/> +as well as on CPAN in the \fIports/\fR directory. +.SS "Acorn RISC OS" +.IX Subsection "Acorn RISC OS" +Because Acorns use ASCII with newlines (\f(CW\*(C`\en\*(C'\fR) in text files as \f(CW\*(C`\e012\*(C'\fR like +Unix, and because Unix filename emulation is turned on by default, +most simple scripts will probably work "out of the box". The native +filesystem is modular, and individual filesystems are free to be +case-sensitive or insensitive, and are usually case-preserving. Some +native filesystems have name length limits, which file and directory +names are silently truncated to fit. Scripts should be aware that the +standard filesystem currently has a name length limit of \fB10\fR +characters, with up to 77 items in a directory, but other filesystems +may not impose such limitations. +.PP +Native filenames are of the form +.PP +.Vb 1 +\& Filesystem#Special_Field::DiskName.$.Directory.Directory.File +.Ve +.PP +where +.PP +.Vb 8 +\& Special_Field is not usually present, but may contain . and $ . +\& Filesystem =~ m|[A\-Za\-z0\-9_]| +\& DsicName =~ m|[A\-Za\-z0\-9_/]| +\& $ represents the root directory +\& . is the path separator +\& @ is the current directory (per filesystem but machine global) +\& ^ is the parent directory +\& Directory and File =~ m|[^\e0\- "\e.\e$\e%\e&:\e@\e\e^\e|\e177]+| +.Ve +.PP +The default filename translation is roughly \f(CW\*(C`tr|/.|./|\*(C'\fR, swapping dots +and slashes. +.PP +Note that \f(CW\*(C`"ADFS::HardDisk.$.File" ne \*(AqADFS::HardDisk.$.File\*(Aq\*(C'\fR and that +the second stage of \f(CW\*(C`$\*(C'\fR interpolation in regular expressions will fall +foul of the \f(CW$.\fR variable if scripts are not careful. +.PP +Logical paths specified by system variables containing comma-separated +search lists are also allowed; hence \f(CW\*(C`System:Modules\*(C'\fR is a valid +filename, and the filesystem will prefix \f(CW\*(C`Modules\*(C'\fR with each section of +\&\f(CW\*(C`System$Path\*(C'\fR until a name is made that points to an object on disk. +Writing to a new file \f(CW\*(C`System:Modules\*(C'\fR would be allowed only if +\&\f(CW\*(C`System$Path\*(C'\fR contains a single item list. The filesystem will also +expand system variables in filenames if enclosed in angle brackets, so +\&\f(CW\*(C`<System$Dir>.Modules\*(C'\fR would look for the file +\&\f(CW\*(C`$ENV{\*(AqSystem$Dir\*(Aq}\ .\ \*(AqModules\*(Aq\*(C'\fR. The obvious implication of this is +that \fBfully qualified filenames can start with \fR\f(CB\*(C`<>\*(C'\fR and the +three-argument form of \f(CW\*(C`open\*(C'\fR should +always be used. +.PP +Because \f(CW\*(C`.\*(C'\fR 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 \f(CW\*(C`.c\*(C'\fR \f(CW\*(C`.h\*(C'\fR \f(CW\*(C`.s\*(C'\fR and \f(CW\*(C`.o\*(C'\fR suffix from +filenames specified in source code and store the respective files in +subdirectories named after the suffix. Hence files are translated: +.PP +.Vb 6 +\& foo.h h.foo +\& C:foo.h C:h.foo (logical path variable) +\& sys/os.h sys.h.os (C compiler groks Unix\-speak) +\& 10charname.c c.10charname +\& 10charname.o o.10charname +\& 11charname_.c c.11charname (assuming filesystem truncates at 10) +.Ve +.PP +The Unix emulation library'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 \fIfoo/bar/baz.h\fR +and \fIfoo/bar/h/baz\fR both map to \fIfoo.bar.h.baz\fR, and that +\&\f(CW\*(C`readdir\*(C'\fR and \f(CW\*(C`glob\*(C'\fR +cannot and do not attempt to emulate the reverse mapping. Other +\&\f(CW\*(C`.\*(C'\fR's in filenames are translated to \f(CW\*(C`/\*(C'\fR. +.PP +As implied above, the environment accessed through +\&\f(CW%ENV\fR is global, and the convention is that program +specific environment variables are of the form \f(CW\*(C`Program$Name\*(C'\fR. +Each filesystem maintains a current directory, +and the current filesystem's current directory is the \fBglobal\fR current +directory. Consequently, sociable programs don't change the current +directory but rely on full pathnames, and programs (and Makefiles) cannot +assume that they can spawn a child process which can change the current +directory without affecting its parent (and everyone else for that +matter). +.PP +Because native operating system filehandles are global and are currently +allocated down from 255, with 0 being a reserved value, the Unix emulation +library emulates Unix filehandles. Consequently, you can't rely on +passing \f(CW\*(C`STDIN\*(C'\fR, \f(CW\*(C`STDOUT\*(C'\fR, or \f(CW\*(C`STDERR\*(C'\fR to your children. +.PP +The desire of users to express filenames of the form +\&\f(CW\*(C`<Foo$Dir>.Bar\*(C'\fR on the command line unquoted causes problems, +too: \f(CW\`\`\fR command output capture has +to perform a guessing game. It assumes that a string \f(CW\*(C`<[^<>]+\e$[^<>]>\*(C'\fR +is a reference to an environment variable, whereas anything else involving +\&\f(CW\*(C`<\*(C'\fR or \f(CW\*(C`>\*(C'\fR is redirection, and generally manages to be 99% +right. Of course, the problem remains that scripts cannot rely on any +Unix tools being available, or that any tools found have Unix-like command +line arguments. +.PP +Extensions and XS are, in theory, buildable by anyone using free +tools. In practice, many don't, as users of the Acorn platform are +used to binary distributions. MakeMaker does run, but no available +make currently copes with MakeMaker's makefiles; even if and when +this should be fixed, the lack of a Unix-like shell will cause +problems with makefile rules, especially lines of the form +\&\f(CW\*(C`cd sdbm && make all\*(C'\fR, and anything using quoting. +.PP +"RISC\ OS" is the proper name for the operating system, but the value +in \f(CW$^O\fR is "riscos" (because we don't like shouting). +.SS "Other perls" +.IX Subsection "Other perls" +Perl has been ported to many platforms that do not fit into any of +the categories listed above. Some, such as AmigaOS, +QNX, Plan 9, and VOS, have been well-integrated into the standard +Perl source code kit. You may need to see the \fIports/\fR directory +on CPAN for information, and possibly binaries, for the likes of: +aos, Atari ST, lynxos, riscos, Novell Netware, Tandem Guardian, +\&\fIetc.\fR (Yes, we know that some of these OSes may fall under the +Unix category, but we are not a standards body.) +.PP +Some approximate operating system names and their \f(CW$^O\fR +values in the "OTHER" category include: +.PP +.Vb 3 +\& OS $^O $Config{archname} +\& \-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\- +\& Amiga DOS amigaos m68k\-amigos +.Ve +.PP +See also: +.IP \(bu 4 +Amiga, \fIREADME.amiga\fR (installed as perlamiga). +.IP \(bu 4 +Plan\ 9, \fIREADME.plan9\fR +.SH "FUNCTION IMPLEMENTATIONS" +.IX Header "FUNCTION IMPLEMENTATIONS" +Listed below are functions that are either completely unimplemented +or else have been implemented differently on various platforms. +Preceding each description will be, in parentheses, a list of +platforms that the description applies to. +.PP +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. +.PP +Be aware, moreover, that even among Unix-ish systems there are variations. +.PP +For many functions, you can also query \f(CW%Config\fR, +exported by default from the \f(CW\*(C`Config\*(C'\fR module. For example, to +check whether the platform has the \f(CW\*(C`lstat\*(C'\fR +call, check \f(CW$Config{d_lstat}\fR. See Config for a +full description of available variables. +.SS "Alphabetical Listing of Perl Functions" +.IX Subsection "Alphabetical Listing of Perl Functions" +.IP \-X 8 +.IX Item "-X" +(Win32) +\&\f(CW\*(C`\-w\*(C'\fR 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). +.Sp +(VMS) +\&\f(CW\*(C`\-r\*(C'\fR, \f(CW\*(C`\-w\*(C'\fR, \f(CW\*(C`\-x\*(C'\fR, and \f(CW\*(C`\-o\*(C'\fR tell whether the file is accessible, +which may not reflect UIC-based file protections. +.Sp +(RISC\ OS) +\&\f(CW\*(C`\-s\*(C'\fR by name on an open file will return the space reserved on disk, +rather than the current extent. \f(CW\*(C`\-s\*(C'\fR on an open filehandle returns the +current size. +.Sp +(Win32, VMS, RISC\ OS) +\&\f(CW\*(C`\-R\*(C'\fR, \f(CW\*(C`\-W\*(C'\fR, \f(CW\*(C`\-X\*(C'\fR, \f(CW\*(C`\-O\*(C'\fR are indistinguishable from \f(CW\*(C`\-r\*(C'\fR, \f(CW\*(C`\-w\*(C'\fR, +\&\f(CW\*(C`\-x\*(C'\fR, \f(CW\*(C`\-o\*(C'\fR. +.Sp +(Win32, VMS, RISC\ OS) +\&\f(CW\*(C`\-g\*(C'\fR, \f(CW\*(C`\-k\*(C'\fR, \f(CW\*(C`\-l\*(C'\fR, \f(CW\*(C`\-u\*(C'\fR, \f(CW\*(C`\-A\*(C'\fR are not particularly meaningful. +.Sp +(Win32) +\&\f(CW\*(C`\-l\*(C'\fR returns true for both symlinks and directory junctions. +.Sp +(VMS, RISC\ OS) +\&\f(CW\*(C`\-p\*(C'\fR is not particularly meaningful. +.Sp +(VMS) +\&\f(CW\*(C`\-d\*(C'\fR is true if passed a device spec without an explicit directory. +.Sp +(Win32) +\&\f(CW\*(C`\-x\*(C'\fR (or \f(CW\*(C`\-X\*(C'\fR) determine if a file ends in one of the executable +suffixes. \f(CW\*(C`\-S\*(C'\fR is meaningless. +.Sp +(RISC\ OS) +\&\f(CW\*(C`\-x\*(C'\fR (or \f(CW\*(C`\-X\*(C'\fR) determine if a file has an executable file type. +.IP alarm 8 +.IX Item "alarm" +(Win32) +Emulated using timers that must be explicitly polled whenever Perl +wants to dispatch "safe signals" and therefore cannot interrupt +blocking system calls. +.IP atan2 8 +.IX Item "atan2" +(Tru64, HP-UX 10.20) +Due to issues with various CPUs, math libraries, compilers, and standards, +results for \f(CW\*(C`atan2\*(C'\fR may vary depending on any combination of the above. +Perl attempts to conform to the Open Group/IEEE standards for the results +returned from \f(CW\*(C`atan2\*(C'\fR, but cannot force the issue if the system Perl is +run on does not allow it. +.Sp +The current version of the standards for \f(CW\*(C`atan2\*(C'\fR is available at +<http://www.opengroup.org/onlinepubs/009695399/functions/atan2.html>. +.IP binmode 8 +.IX Item "binmode" +(RISC\ OS) +Meaningless. +.Sp +(VMS) +Reopens file and restores pointer; if function fails, underlying +filehandle may be closed, or pointer may be in a different position. +.Sp +(Win32) +The value returned by \f(CW\*(C`tell\*(C'\fR may be affected +after the call, and the filehandle may be flushed. +.IP chdir 8 +.IX Item "chdir" +(Win32) +The current directory reported by the system may include any symbolic +links specified to \fBchdir()\fR. +.IP chmod 8 +.IX Item "chmod" +(Win32) +Only good for changing "owner" read-write access; "group" and "other" +bits are meaningless. +.Sp +(RISC\ OS) +Only good for changing "owner" and "other" read-write access. +.Sp +(VOS) +Access permissions are mapped onto VOS access-control list changes. +.Sp +(Cygwin) +The actual permissions set depend on the value of the \f(CW\*(C`CYGWIN\*(C'\fR variable +in the SYSTEM environment settings. +.Sp +(Android) +Setting the exec bit on some locations (generally \fI/sdcard\fR) will return true +but not actually set the bit. +.Sp +(VMS) +A mode argument of zero sets permissions to the user's default permission mask +rather than disabling all permissions. +.IP chown 8 +.IX Item "chown" +(Plan\ 9, RISC\ OS) +Not implemented. +.Sp +(Win32) +Does nothing, but won't fail. +.Sp +(VOS) +A little funky, because VOS's notion of ownership is a little funky. +.IP chroot 8 +.IX Item "chroot" +(Win32, VMS, Plan\ 9, RISC\ OS, VOS) +Not implemented. +.IP crypt 8 +.IX Item "crypt" +(Win32) +May not be available if library or source was not provided when building +perl. +.Sp +(Android) +Not implemented. +.IP dbmclose 8 +.IX Item "dbmclose" +(VMS, Plan\ 9, VOS) +Not implemented. +.IP dbmopen 8 +.IX Item "dbmopen" +(VMS, Plan\ 9, VOS) +Not implemented. +.IP dump 8 +.IX Item "dump" +(RISC\ OS) +Not useful. +.Sp +(Cygwin, Win32) +Not supported. +.Sp +(VMS) +Invokes VMS debugger. +.IP exec 8 +.IX Item "exec" +(Win32) +\&\f(CW\*(C`exec LIST\*(C'\fR without the use of indirect object syntax (\f(CW\*(C`exec PROGRAM LIST\*(C'\fR) +may fall back to trying the shell if the first \f(CWspawn()\fR fails. +.Sp +Note that the list form of \fBexec()\fR is emulated since the Win32 API +\&\fBCreateProcess()\fR accepts a simple string rather than an array of +command-line arguments. This may have security implications for your +code. +.Sp +(SunOS, Solaris, HP-UX) +Does not automatically flush output handles on some platforms. +.IP exit 8 +.IX Item "exit" +(VMS) +Emulates Unix \f(CW\*(C`exit\*(C'\fR (which considers \f(CW\*(C`exit 1\*(C'\fR to indicate an error) by +mapping the \f(CW1\fR to \f(CW\*(C`SS$_ABORT\*(C'\fR (\f(CW44\fR). This behavior may be overridden +with the pragma \f(CW\*(C`use vmsish \*(Aqexit\*(Aq\*(C'\fR. As with +the CRTL's \f(CWexit()\fR function, \f(CW\*(C`exit 0\*(C'\fR is also mapped to an exit status +of \f(CW\*(C`SS$_NORMAL\*(C'\fR (\f(CW1\fR); this mapping cannot be overridden. Any other +argument to \f(CW\*(C`exit\*(C'\fR +is used directly as Perl'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. +.Sp +(Solaris) +\&\f(CW\*(C`exit\*(C'\fR resets file pointers, which is a problem when called +from a child process (created by \f(CW\*(C`fork\*(C'\fR) in +\&\f(CW\*(C`BEGIN\*(C'\fR. +A workaround is to use \f(CW\*(C`POSIX::_exit\*(C'\fR. +.Sp +.Vb 3 +\& exit unless $Config{archname} =~ /\ebsolaris\eb/; +\& require POSIX; +\& POSIX::_exit(0); +.Ve +.IP fcntl 8 +.IX Item "fcntl" +(Win32) +Not implemented. +.Sp +(VMS) +Some functions available based on the version of VMS. +.IP flock 8 +.IX Item "flock" +(VMS, RISC\ OS, VOS) +Not implemented. +.IP fork 8 +.IX Item "fork" +(AmigaOS, RISC\ OS, VMS) +Not implemented. +.Sp +(Win32) +Emulated using multiple interpreters. See perlfork. +.Sp +(SunOS, Solaris, HP-UX) +Does not automatically flush output handles on some platforms. +.IP getlogin 8 +.IX Item "getlogin" +(RISC\ OS) +Not implemented. +.IP getpgrp 8 +.IX Item "getpgrp" +(Win32, VMS, RISC\ OS) +Not implemented. +.IP getppid 8 +.IX Item "getppid" +(Win32, RISC\ OS) +Not implemented. +.IP getpriority 8 +.IX Item "getpriority" +(Win32, VMS, RISC\ OS, VOS) +Not implemented. +.IP getpwnam 8 +.IX Item "getpwnam" +(Win32) +Not implemented. +.Sp +(RISC\ OS) +Not useful. +.IP getgrnam 8 +.IX Item "getgrnam" +(Win32, VMS, RISC\ OS) +Not implemented. +.IP getnetbyname 8 +.IX Item "getnetbyname" +(Android, Win32, Plan\ 9) +Not implemented. +.IP getpwuid 8 +.IX Item "getpwuid" +(Win32) +Not implemented. +.Sp +(RISC\ OS) +Not useful. +.IP getgrgid 8 +.IX Item "getgrgid" +(Win32, VMS, RISC\ OS) +Not implemented. +.IP getnetbyaddr 8 +.IX Item "getnetbyaddr" +(Android, Win32, Plan\ 9) +Not implemented. +.IP getprotobynumber 8 +.IX Item "getprotobynumber" +(Android) +Not implemented. +.IP getpwent 8 +.IX Item "getpwent" +(Android, Win32) +Not implemented. +.IP getgrent 8 +.IX Item "getgrent" +(Android, Win32, VMS) +Not implemented. +.IP gethostbyname 8 +.IX Item "gethostbyname" +(Irix\ 5) +\&\f(CWgethostbyname(\*(Aqlocalhost\*(Aq)\fR does not work everywhere: you may have +to use \f(CWgethostbyname(\*(Aq127.0.0.1\*(Aq)\fR. +.IP gethostent 8 +.IX Item "gethostent" +(Win32) +Not implemented. +.IP getnetent 8 +.IX Item "getnetent" +(Android, Win32, Plan\ 9) +Not implemented. +.IP getprotoent 8 +.IX Item "getprotoent" +(Android, Win32, Plan\ 9) +Not implemented. +.IP getservent 8 +.IX Item "getservent" +(Win32, Plan\ 9) +Not implemented. +.IP seekdir 8 +.IX Item "seekdir" +(Android) +Not implemented. +.IP sethostent 8 +.IX Item "sethostent" +(Android, Win32, Plan\ 9, RISC\ OS) +Not implemented. +.IP setnetent 8 +.IX Item "setnetent" +(Win32, Plan\ 9, RISC\ OS) +Not implemented. +.IP setprotoent 8 +.IX Item "setprotoent" +(Android, Win32, Plan\ 9, RISC\ OS) +Not implemented. +.IP setservent 8 +.IX Item "setservent" +(Plan\ 9, Win32, RISC\ OS) +Not implemented. +.IP endpwent 8 +.IX Item "endpwent" +(Win32) +Not implemented. +.Sp +(Android) +Either not implemented or a no-op. +.IP endgrent 8 +.IX Item "endgrent" +(Android, RISC\ OS, VMS, Win32) +Not implemented. +.IP endhostent 8 +.IX Item "endhostent" +(Android, Win32) +Not implemented. +.IP endnetent 8 +.IX Item "endnetent" +(Android, Win32, Plan\ 9) +Not implemented. +.IP endprotoent 8 +.IX Item "endprotoent" +(Android, Win32, Plan\ 9) +Not implemented. +.IP endservent 8 +.IX Item "endservent" +(Plan\ 9, Win32) +Not implemented. +.IP getsockopt 8 +.IX Item "getsockopt" +(Plan\ 9) +Not implemented. +.IP glob 8 +.IX Item "glob" +This operator is implemented via the \f(CW\*(C`File::Glob\*(C'\fR extension +on most platforms. See File::Glob for portability information. +.IP gmtime 8 +.IX Item "gmtime" +In theory, \f(CW\*(C`gmtime\*(C'\fR is reliable from \-2**63 to 2**63\-1. However, +because work-arounds in the implementation use floating point numbers, +it will become inaccurate as the time gets larger. This is a bug and +will be fixed in the future. +.Sp +(VOS) +Time values are 32\-bit quantities. +.IP ioctl 8 +.IX Item "ioctl" +(VMS) +Not implemented. +.Sp +(Win32) +Available only for socket handles, and it does what the \f(CWioctlsocket()\fR call +in the Winsock API does. +.Sp +(RISC\ OS) +Available only for socket handles. +.IP kill 8 +.IX Item "kill" +(RISC\ OS) +Not implemented, hence not useful for taint checking. +.Sp +(Win32) +\&\f(CW\*(C`kill\*(C'\fR doesn't send a signal to the identified process like it does on +Unix platforms. Instead \f(CW\*(C`kill($sig, $pid)\*(C'\fR terminates the process +identified by \f(CW$pid\fR, and makes it exit immediately with exit status +\&\f(CW$sig\fR. As in Unix, if \f(CW$sig\fR is 0 and the specified process exists, it +returns true without actually terminating it. +.Sp +(Win32) +\&\f(CW\*(C`kill(\-9, $pid)\*(C'\fR will terminate the process specified by \f(CW$pid\fR 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 +\&\f(CW$pid\fR. +.Sp +(VMS) +A pid of \-1 indicating all processes on the system is not currently +supported. +.IP link 8 +.IX Item "link" +(RISC\ OS, VOS) +Not implemented. +.Sp +(AmigaOS) +Link count not updated because hard links are not quite that hard +(They are sort of half-way between hard and soft links). +.Sp +(Win32) +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. +.Sp +(VMS) +Available on 64 bit OpenVMS 8.2 and later. +.IP localtime 8 +.IX Item "localtime" +\&\f(CW\*(C`localtime\*(C'\fR has the same range as "gmtime", but because time zone +rules change, its accuracy for historical and future times may degrade +but usually by no more than an hour. +.IP lstat 8 +.IX Item "lstat" +(RISC\ OS) +Not implemented. +.Sp +(Win32) +Treats directory junctions as symlinks. +.IP msgctl 8 +.IX Item "msgctl" +.PD 0 +.IP msgget 8 +.IX Item "msgget" +.IP msgsnd 8 +.IX Item "msgsnd" +.IP msgrcv 8 +.IX Item "msgrcv" +.PD +(Android, Win32, VMS, Plan\ 9, RISC\ OS, VOS) +Not implemented. +.IP open 8 +.IX Item "open" +(RISC\ OS) +Open modes \f(CW\*(C`|\-\*(C'\fR and \f(CW\*(C`\-|\*(C'\fR are unsupported. +.Sp +(SunOS, Solaris, HP-UX) +Opening a process does not automatically flush output handles on some +platforms. +.Sp +(Win32) +Both of modes \f(CW\*(C`|\-\*(C'\fR and \f(CW\*(C`\-|\*(C'\fR are supported, but the list form is +emulated since the Win32 API \fBCreateProcess()\fR accepts a simple string +rather than an array of arguments. This may have security +implications for your code. +.IP readlink 8 +.IX Item "readlink" +(VMS, RISC\ OS) +Not implemented. +.Sp +(Win32) +\&\fBreadlink()\fR on a directory junction returns the object name, not a +simple path. +.IP rename 8 +.IX Item "rename" +(Win32) +Can't move directories between directories on different logical volumes. +.IP rewinddir 8 +.IX Item "rewinddir" +(Win32) +Will not cause \f(CW\*(C`readdir\*(C'\fR to re-read the +directory stream. The entries already read before the \f(CW\*(C`rewinddir\*(C'\fR call +will just be returned again from a cache buffer. +.IP select 8 +.IX Item "select" +(Win32, VMS) +Only implemented on sockets. +.Sp +(RISC\ OS) +Only reliable on sockets. +.Sp +Note that the \f(CW\*(C`select FILEHANDLE\*(C'\fR form is +generally portable. +.IP semctl 8 +.IX Item "semctl" +.PD 0 +.IP semget 8 +.IX Item "semget" +.IP semop 8 +.IX Item "semop" +.PD +(Android, Win32, VMS, RISC\ OS) +Not implemented. +.IP setgrent 8 +.IX Item "setgrent" +(Android, VMS, Win32, RISC\ OS) +Not implemented. +.IP setpgrp 8 +.IX Item "setpgrp" +(Win32, VMS, RISC\ OS, VOS) +Not implemented. +.IP setpriority 8 +.IX Item "setpriority" +(Win32, VMS, RISC\ OS, VOS) +Not implemented. +.IP setpwent 8 +.IX Item "setpwent" +(Android, Win32, RISC\ OS) +Not implemented. +.IP setsockopt 8 +.IX Item "setsockopt" +(Plan\ 9) +Not implemented. +.IP shmctl 8 +.IX Item "shmctl" +.PD 0 +.IP shmget 8 +.IX Item "shmget" +.IP shmread 8 +.IX Item "shmread" +.IP shmwrite 8 +.IX Item "shmwrite" +.PD +(Android, Win32, VMS, RISC\ OS) +Not implemented. +.IP sleep 8 +.IX Item "sleep" +(Win32) +Emulated using synchronization functions such that it can be +interrupted by \f(CW\*(C`alarm\*(C'\fR, and limited to a +maximum of 4294967 seconds, approximately 49 days. +.IP socketpair 8 +.IX Item "socketpair" +(RISC\ OS) +Not implemented. +.Sp +(VMS) +Available on 64 bit OpenVMS 8.2 and later. +.IP stat 8 +.IX Item "stat" +Platforms that do not have \f(CW\*(C`rdev\*(C'\fR, \f(CW\*(C`blksize\*(C'\fR, or \f(CW\*(C`blocks\*(C'\fR will return +these as \f(CW\*(Aq\*(Aq\fR, so numeric comparison or manipulation of these fields may +cause 'not numeric' warnings. +.Sp +(Mac\ OS\ X) +\&\f(CW\*(C`ctime\*(C'\fR not supported on UFS. +.Sp +(Win32) +\&\f(CW\*(C`ctime\*(C'\fR is creation time instead of inode change time. +.Sp +(VMS) +\&\f(CW\*(C`dev\*(C'\fR and \f(CW\*(C`ino\*(C'\fR are not necessarily reliable. +.Sp +(RISC\ OS) +\&\f(CW\*(C`mtime\*(C'\fR, \f(CW\*(C`atime\*(C'\fR and \f(CW\*(C`ctime\*(C'\fR all return the last modification time. +\&\f(CW\*(C`dev\*(C'\fR and \f(CW\*(C`ino\*(C'\fR are not necessarily reliable. +.Sp +(OS/2) +\&\f(CW\*(C`dev\*(C'\fR, \f(CW\*(C`rdev\*(C'\fR, \f(CW\*(C`blksize\*(C'\fR, and \f(CW\*(C`blocks\*(C'\fR are not available. \f(CW\*(C`ino\*(C'\fR is not +meaningful and will differ between stat calls on the same file. +.Sp +(Cygwin) +Some versions of cygwin when doing a \f(CWstat("foo")\fR and not finding it +may then attempt to \f(CWstat("foo.exe")\fR. +.IP symlink 8 +.IX Item "symlink" +(RISC\ OS) +Not implemented. +.Sp +(Win32) +Requires either elevated permissions or developer mode and a +sufficiently recent version of Windows 10. You can check whether the current +process has the required privileges using the +\&\fBWin32::IsSymlinkCreationAllowed()\fR +function. +.Sp +Since Windows needs to know whether the target is a directory or not when +creating the link the target Perl will only create the link as a directory +link when the target exists and is a directory. +.Sp +Windows does not recognize forward slashes as path separators in +symbolic links. Hence on Windows, any \f(CW\*(C`/\*(C'\fR in the \fIOLDFILE\fR +parameter to \fBsymlink()\fR are converted to \f(CW\*(C`\e\*(C'\fR. This is reflected in +the result returned by \fBreadlink()\fR, the \f(CW\*(C`\e\*(C'\fR in the result are not +converted back to \f(CW\*(C`/\*(C'\fR. +.Sp +(VMS) +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. +.IP syscall 8 +.IX Item "syscall" +(Win32, VMS, RISC\ OS, VOS) +Not implemented. +.IP sysopen 8 +.IX Item "sysopen" +(Mac\ OS, OS/390) +The traditional \f(CW0\fR, \f(CW1\fR, and \f(CW2\fR MODEs are implemented with different +numeric values on some systems. The flags exported by \f(CW\*(C`Fcntl\*(C'\fR +(\f(CW\*(C`O_RDONLY\*(C'\fR, \f(CW\*(C`O_WRONLY\*(C'\fR, \f(CW\*(C`O_RDWR\*(C'\fR) should work everywhere though. +.IP system 8 +.IX Item "system" +(Win32) +As an optimization, may not call the command shell specified in +\&\f(CW$ENV{PERL5SHELL}\fR. \f(CW\*(C`system(1, @args)\*(C'\fR spawns an external +process and immediately returns its process designator, without +waiting for it to terminate. Return value may be used subsequently +in \f(CW\*(C`wait\*(C'\fR or \f(CW\*(C`waitpid\*(C'\fR. +Failure to \f(CWspawn()\fR a subprocess is indicated by setting +\&\f(CW$?\fR to \f(CW\*(C`255 << 8\*(C'\fR. \f(CW$?\fR is set in a +way compatible with Unix (i.e. the exit status of the subprocess is +obtained by \f(CW\*(C`$? >> 8\*(C'\fR, as described in the documentation). +.Sp +Note that the list form of \fBsystem()\fR is emulated since the Win32 API +\&\fBCreateProcess()\fR accepts a simple string rather than an array of +command-line arguments. This may have security implications for your +code. +.Sp +(RISC\ OS) +There is no shell to process metacharacters, and the native standard is +to pass a command line terminated by "\en" "\er" or "\e0" to the spawned +program. Redirection such as \f(CW\*(C`> foo\*(C'\fR is performed (if at all) by +the run time library of the spawned program. \f(CW\*(C`system LIST\*(C'\fR will call +the Unix emulation library's \f(CW\*(C`exec\*(C'\fR emulation, +which attempts to provide emulation of the stdin, stdout, stderr in force +in the parent, provided the child program uses a compatible version of the +emulation library. \f(CW\*(C`system SCALAR\*(C'\fR will call the native command line +directly and no such emulation of a child Unix program will occur. +Mileage \fBwill\fR vary. +.Sp +(Win32) +\&\f(CW\*(C`system LIST\*(C'\fR without the use of indirect object syntax (\f(CW\*(C`system PROGRAM LIST\*(C'\fR) +may fall back to trying the shell if the first \f(CWspawn()\fR fails. +.Sp +(SunOS, Solaris, HP-UX) +Does not automatically flush output handles on some platforms. +.Sp +(VMS) +As with Win32, \f(CW\*(C`system(1, @args)\*(C'\fR spawns an external process and +immediately returns its process designator without waiting for the +process to terminate. In this case the return value may be used subsequently +in \f(CW\*(C`wait\*(C'\fR or \f(CW\*(C`waitpid\*(C'\fR. +Otherwise 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 +\&\f(CW\*(C`use vmsish \*(Aqstatus\*(Aq\*(C'\fR). 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 +"$?" in perlvms. +.IP telldir 8 +.IX Item "telldir" +(Android) +Not implemented. +.IP times 8 +.IX Item "times" +(Win32) +"Cumulative" times will be bogus. On anything other than Windows NT +or Windows 2000, "system" time will be bogus, and "user" time is +actually the time returned by the \f(CWclock()\fR function in the C +runtime library. +.Sp +(RISC\ OS) +Not useful. +.IP truncate 8 +.IX Item "truncate" +(Older versions of VMS) +Not implemented. +.Sp +(VOS) +Truncation to same-or-shorter lengths only. +.Sp +(Win32) +If a FILEHANDLE is supplied, it must be writable and opened in append +mode (i.e., use \f(CW\*(C`open(my $fh, \*(Aq>>\*(Aq, \*(Aqfilename\*(Aq)\*(C'\fR +or \f(CW\*(C`sysopen(my $fh, ..., O_APPEND|O_RDWR)\*(C'\fR. If a filename is supplied, it +should not be held open elsewhere. +.IP umask 8 +.IX Item "umask" +Returns \f(CW\*(C`undef\*(C'\fR where unavailable. +.Sp +(AmigaOS) +\&\f(CW\*(C`umask\*(C'\fR works but the correct permissions are set only when the file +is finally closed. +.IP utime 8 +.IX Item "utime" +(VMS, RISC\ OS) +Only the modification time is updated. +.Sp +(Win32) +May not behave as expected. Behavior depends on the C runtime +library's implementation of \f(CWutime()\fR, and the filesystem +being used. The FAT filesystem typically does not support an "access +time" field, and it may limit timestamps to a granularity of two seconds. +.IP wait 8 +.IX Item "wait" +.PD 0 +.IP waitpid 8 +.IX Item "waitpid" +.PD +(Win32) +Can only be applied to process handles returned for processes spawned +using \f(CW\*(C`system(1, ...)\*(C'\fR or pseudo processes created with +\&\f(CW\*(C`fork\*(C'\fR. +.Sp +(RISC\ OS) +Not useful. +.SH "Supported Platforms" +.IX Header "Supported Platforms" +The following platforms are known to build Perl 5.12 (as of April 2010, +its release date) from the standard source code distribution available +at <http://www.cpan.org/src> +.IP "Linux (x86, ARM, IA64)" 4 +.IX Item "Linux (x86, ARM, IA64)" +.PD 0 +.IP HP-UX 4 +.IX Item "HP-UX" +.IP AIX 4 +.IX Item "AIX" +.IP Win32 4 +.IX Item "Win32" +.RS 4 +.IP "Windows 2000" 4 +.IX Item "Windows 2000" +.IP "Windows XP" 4 +.IX Item "Windows XP" +.IP "Windows Server 2003" 4 +.IX Item "Windows Server 2003" +.IP "Windows Vista" 4 +.IX Item "Windows Vista" +.IP "Windows Server 2008" 4 +.IX Item "Windows Server 2008" +.IP "Windows 7" 4 +.IX Item "Windows 7" +.RE +.RS 4 +.RE +.IP Cygwin 4 +.IX Item "Cygwin" +.PD +Some tests are known to fail: +.RS 4 +.IP \(bu 4 +\&\fIext/XS\-APItest/t/call_checker.t\fR \- see +<https://github.com/Perl/perl5/issues/10750> +.IP \(bu 4 +\&\fIdist/I18N\-Collate/t/I18N\-Collate.t\fR +.IP \(bu 4 +\&\fIext/Win32CORE/t/win32core.t\fR \- may fail on recent cygwin installs. +.RE +.RS 4 +.RE +.IP "Solaris (x86, SPARC)" 4 +.IX Item "Solaris (x86, SPARC)" +.PD 0 +.IP OpenVMS 4 +.IX Item "OpenVMS" +.RS 4 +.IP "Alpha (7.2 and later)" 4 +.IX Item "Alpha (7.2 and later)" +.IP "I64 (8.2 and later)" 4 +.IX Item "I64 (8.2 and later)" +.RE +.RS 4 +.RE +.IP NetBSD 4 +.IX Item "NetBSD" +.IP FreeBSD 4 +.IX Item "FreeBSD" +.IP "Debian GNU/kFreeBSD" 4 +.IX Item "Debian GNU/kFreeBSD" +.IP Haiku 4 +.IX Item "Haiku" +.IP "Irix (6.5. What else?)" 4 +.IX Item "Irix (6.5. What else?)" +.IP OpenBSD 4 +.IX Item "OpenBSD" +.IP "Dragonfly BSD" 4 +.IX Item "Dragonfly BSD" +.IP "Midnight BSD" 4 +.IX Item "Midnight BSD" +.IP "QNX Neutrino RTOS (6.5.0)" 4 +.IX Item "QNX Neutrino RTOS (6.5.0)" +.IP "MirOS BSD" 4 +.IX Item "MirOS BSD" +.IP "Stratus OpenVOS (17.0 or later)" 4 +.IX Item "Stratus OpenVOS (17.0 or later)" +.PD +Caveats: +.RS 4 +.IP "time_t issues that may or may not be fixed" 4 +.IX Item "time_t issues that may or may not be fixed" +.RE +.RS 4 +.RE +.PD 0 +.IP "Stratus VOS / OpenVOS" 4 +.IX Item "Stratus VOS / OpenVOS" +.IP AIX 4 +.IX Item "AIX" +.IP Android 4 +.IX Item "Android" +.IP FreeMINT 4 +.IX Item "FreeMINT" +.PD +Perl now builds with FreeMiNT/Atari. It fails a few tests, that needs +some investigation. +.Sp +The FreeMiNT port uses GNU dld for loadable module capabilities. So +ensure you have that library installed when building perl. +.SH "EOL Platforms" +.IX Header "EOL Platforms" +.SS "(Perl 5.37.1)" +.IX Subsection "(Perl 5.37.1)" +The following platforms were supported by a previous version of +Perl but have been officially removed from Perl's source code +as of 5.37.1: +.IP Ultrix 4 +.IX Item "Ultrix" +.SS "(Perl 5.36)" +.IX Subsection "(Perl 5.36)" +The following platforms were supported by a previous version of +Perl but have been officially removed from Perl's source code +as of 5.36: +.IP NetWare 4 +.IX Item "NetWare" +.PD 0 +.IP DOS/DJGPP 4 +.IX Item "DOS/DJGPP" +.IP "AT&T UWIN" 4 +.IX Item "AT&T UWIN" +.PD +.SS "(Perl 5.20)" +.IX Subsection "(Perl 5.20)" +The following platforms were supported by a previous version of +Perl but have been officially removed from Perl's source code +as of 5.20: +.IP "AT&T 3b1" 4 +.IX Item "AT&T 3b1" +.SS "(Perl 5.14)" +.IX Subsection "(Perl 5.14)" +The following platforms were supported up to 5.10. They may still +have worked in 5.12, but supporting code has been removed for 5.14: +.IP "Windows 95" 4 +.IX Item "Windows 95" +.PD 0 +.IP "Windows 98" 4 +.IX Item "Windows 98" +.IP "Windows ME" 4 +.IX Item "Windows ME" +.IP "Windows NT4" 4 +.IX Item "Windows NT4" +.PD +.SS "(Perl 5.12)" +.IX Subsection "(Perl 5.12)" +The following platforms were supported by a previous version of +Perl but have been officially removed from Perl's source code +as of 5.12: +.IP "Atari MiNT" 4 +.IX Item "Atari MiNT" +.PD 0 +.IP "Apollo Domain/OS" 4 +.IX Item "Apollo Domain/OS" +.IP "Apple Mac OS 8/9" 4 +.IX Item "Apple Mac OS 8/9" +.IP "Tenon Machten" 4 +.IX Item "Tenon Machten" +.PD +.SH "Supported Platforms (Perl 5.8)" +.IX Header "Supported Platforms (Perl 5.8)" +As of July 2002 (the Perl release 5.8.0), the following platforms were +able to build Perl from the standard source code distribution +available at <http://www.cpan.org/src/> +.PP +.Vb 10 +\& AIX +\& BeOS +\& BSD/OS (BSDi) +\& Cygwin +\& DG/UX +\& DOS DJGPP 1) +\& DYNIX/ptx +\& EPOC R5 +\& FreeBSD +\& HI\-UXMPP (Hitachi) (5.8.0 worked but we didn\*(Aqt know it) +\& HP\-UX +\& IRIX +\& Linux +\& Mac OS Classic +\& Mac OS X (Darwin) +\& MPE/iX +\& NetBSD +\& NetWare +\& NonStop\-UX +\& ReliantUNIX (formerly SINIX) +\& OpenBSD +\& OpenVMS (formerly VMS) +\& Open UNIX (Unixware) (since Perl 5.8.1/5.9.0) +\& OS/2 +\& OS/400 (using the PASE) (since Perl 5.8.1/5.9.0) +\& POSIX\-BC (formerly BS2000) +\& QNX +\& Solaris +\& SunOS 4 +\& SUPER\-UX (NEC) +\& Tru64 UNIX (formerly DEC OSF/1, Digital UNIX) +\& UNICOS +\& UNICOS/mk +\& UTS +\& VOS / OpenVOS +\& Win95/98/ME/2K/XP 2) +\& WinCE +\& z/OS (formerly OS/390) +\& VM/ESA +\& +\& 1) in DOS mode either the DOS or OS/2 ports can be used +\& 2) compilers: Borland, MinGW (GCC), VC6 +.Ve +.PP +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. +.PP +.Vb 10 +\& BSD/OS +\& DomainOS +\& Hurd +\& LynxOS +\& MachTen +\& PowerMAX +\& SCO SV +\& SVR4 +\& Unixware +\& Windows 3.1 +.Ve +.PP +Known to be broken for 5.8.0 (but 5.6.1 and 5.7.2 can be used): +.PP +.Vb 1 +\& AmigaOS 3 +.Ve +.PP +The following platforms have been known to build Perl from source in +the past (5.005_03 and earlier), but we haven't been able to verify +their status for the current release, either because the +hardware/software platforms are rare or because we don't have an +active champion on these platforms\-\-or both. They used to work, +though, so go ahead and try compiling them, and let +<https://github.com/Perl/perl5/issues> know +of any trouble. +.PP +.Vb 10 +\& 3b1 +\& A/UX +\& ConvexOS +\& CX/UX +\& DC/OSx +\& DDE SMES +\& DOS EMX +\& Dynix +\& EP/IX +\& ESIX +\& FPS +\& GENIX +\& Greenhills +\& ISC +\& MachTen 68k +\& MPC +\& NEWS\-OS +\& NextSTEP +\& OpenSTEP +\& Opus +\& Plan 9 +\& RISC/os +\& SCO ODT/OSR +\& Stellar +\& SVR2 +\& TI1500 +\& TitanOS +\& Unisys Dynix +.Ve +.PP +The following platforms have their own source code distributions and +binaries available via <http://www.cpan.org/ports/> +.PP +.Vb 1 +\& Perl release +\& +\& OS/400 (ILE) 5.005_02 +\& Tandem Guardian 5.004 +.Ve +.PP +The following platforms have only binaries available via +<http://www.cpan.org/ports/index.html> : +.PP +.Vb 1 +\& Perl release +\& +\& Acorn RISCOS 5.005_02 +\& AOS 5.002 +\& LynxOS 5.004_02 +.Ve +.PP +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 +<http://www.cpan.org/ports/index.html> for binary distributions. +.SH "SEE ALSO" +.IX Header "SEE ALSO" +perlaix, perlamiga, perlbs2000, +perlcygwin, +perlebcdic, perlfreebsd, perlhurd, perlhpux, perlirix, +perlmacosx, +perlos2, perlos390, perlos400, +perlplan9, perlqnx, perlsolaris, perltru64, +perlunicode, perlvms, perlvos, perlwin32, and Win32. +.SH "AUTHORS / CONTRIBUTORS" +.IX Header "AUTHORS / CONTRIBUTORS" +Abigail <abigail@abigail.be>, +Charles Bailey <bailey@newman.upenn.edu>, +Graham Barr <gbarr@pobox.com>, +Tom Christiansen <tchrist@perl.com>, +Nicholas Clark <nick@ccl4.org>, +Thomas Dorner <Thomas.Dorner@start.de>, +Andy Dougherty <doughera@lafayette.edu>, +Dominic Dunlop <domo@computer.org>, +Neale Ferguson <neale@vma.tabnsw.com.au>, +David J. Fiander <davidf@mks.com>, +Paul Green <Paul.Green@stratus.com>, +M.J.T. Guy <mjtg@cam.ac.uk>, +Jarkko Hietaniemi <jhi@iki.fi>, +Luther Huffman <lutherh@stratcom.com>, +Nick Ing-Simmons <nick@ing\-simmons.net>, +Andreas J. König <a.koenig@mind.de>, +Markus Laker <mlaker@contax.co.uk>, +Andrew M. Langmead <aml@world.std.com>, +Lukas Mai <l.mai@web.de>, +Larry Moore <ljmoore@freespace.net>, +Paul Moore <Paul.Moore@uk.origin\-it.com>, +Chris Nandor <pudge@pobox.com>, +Matthias Neeracher <neeracher@mac.com>, +Philip Newton <pne@cpan.org>, +Gary Ng <71564.1743@CompuServe.COM>, +Tom Phoenix <rootbeer@teleport.com>, +André Pirard <A.Pirard@ulg.ac.be>, +Peter Prymmer <pvhp@forte.com>, +Hugo van der Sanden <hv@crypt0.demon.co.uk>, +Gurusamy Sarathy <gsar@activestate.com>, +Paul J. Schinder <schinder@pobox.com>, +Michael G Schwern <schwern@pobox.com>, +Dan Sugalski <dan@sidhe.org>, +Nathan Torkington <gnat@frii.com>, +John Malmberg <wb8tyw@qsl.net> |