input is the file to read for POD source (the POD can be embedded in code). If input isn't given, it defaults to STDIN. output, if given, is the file to which to write the formatted output. If output isn't given, the formatted output is written to STDOUT. Several POD files can be processed in the same pod2man invocation (saving module load and compile times) by providing multiple pairs of input and output files on the command line.
--section, --release, --center, --date, and --official can be used to set the headers and footers to use; if not given, Pod::Man will assume various defaults. See below or Pod::Man for details.
pod2man assumes that your *roff formatters have a fixed-width font named CW. If yours is called something else (like CR), use --fixed to specify it. This generally only matters for troff output for printing. Similarly, you can set the fonts used for bold, italic, and bold italic fixed-width output.
Besides the obvious pod conversions, Pod::Man, and therefore pod2man also takes care of formatting func(), func(n), and simple variable references like $foo or @bar so you don't have to use code escapes for them; complex expressions like $fred{'stuff'} will still need to be escaped, though. It also translates dashes that aren't used as hyphens into en dashes, makes long dashes---like this---into proper em dashes, fixes ``paired quotes,'' and takes care of several other troff-specific tweaks. See Pod::Man for complete information.
Note that this option is probably not useful when converting multiple POD files at once. The convention for Unix man pages for commands is for the man page title to be in all-uppercase even if the command isn't.
quotes may also be set to the special value "none", in which case no quote marks are added around C<> text (but the font is still changed for troff output).
By default, section 1 will be used unless the file ends in .pm in which case section 3 will be selected.
pod2man program > program.1
pod2man SomeModule.pm /usr/perl/man/man3/SomeModule.3
pod2man --section=7 note.pod > note.7
If you would like to print out a lot of man page continuously, you probably want to set the C and D registers to set contiguous page numbering and even/odd paging, at least on some versions of man(7).
troff -man -rC1 -rD1 perl.1 perldata.1 perlsyn.1 ...
To get index entries on stderr, turn on the F register, as in:
troff -man -rF1 perl.1
The indexing merely outputs messages via ".tm" for each major page, section, subsection, item, and any "X<>" directives. See Pod::Man for more details.
The name of the program being documented is conventionally written in bold (using B<>) wherever it occurs, as are all program options. Arguments should be written in italics (I<>). Functions are traditionally written in italics; if you write a function as function(), Pod::Man will take care of this for you. Literal code or commands should be in C<>. References to other man pages should be in the form "manpage(section)", and Pod::Man will automatically format those appropriately. As an exception, it's traditional not to use this form when referring to module documentation; use "L<Module::Name>" instead.
References to other programs or functions are normally in the form of man page references so that cross-referencing tools can provide the user with links and the like. It's possible to overdo this, though, so be careful not to clutter your documentation with too much markup.
The major headers should be set out using a "=head1" directive, and are historically written in the rather startling ALL UPPER CASE format, although this is not mandatory. Minor headers may be included using "=head2", and are typically in mixed case.
The standard sections of a manual page are:
foo, bar - programs to do something
Manual page indexers are often extremely picky about the format of this section, so don't put anything in it except this line. A single dash, and only a single dash, should separate the list of programs or functions from the description. Functions should not be qualified with "()" or the like. The description should ideally fit on a single line, even if a man program replaces the dash with a few tabs.
=head2 Normal Usage
=head2 Advanced Features
=head2 Writing Configuration Files
or whatever is appropriate for your documentation.
=item B<--section>=I<manext>
Synonymous options (like both the short and long forms) are separated by a comma and a space on the same "=item" line, or optionally listed as their own item with a reference to the canonical name. For example, since --section can also be written as -s, the above would be:
=item B<-s> I<manext>, B<--section>=I<manext>
(Writing the short option first is arguably easier to read, since the long option is long enough to draw the eye to it anyway and the short option can otherwise get lost in visual noise.)
If applicable, please include details on what the user should do to correct the error; documenting an error as indicating ``the input buffer is too small'' without telling the user how to increase the size of the input buffer (or at least telling them that it isn't possible) aren't very useful.
Don't just present an example without explaining what it does. Adding a short paragraph saying what the example will do can increase the value of the example immensely.
=over 6
=item HOME
Used to determine the user's home directory. F<.foorc> in this
directory is read for configuration details, if it exists.
=back
Since environment variables are normally in all uppercase, no additional special formatting is generally needed; they're glaring enough as it is.
Copyright YEAR(s) by YOUR NAME(s)
(No, (C) is not needed. No, ``all rights reserved'' is not needed.)
For licensing the easiest way is to use the same licensing as Perl itself:
This library is free software; you may redistribute it and/or modify
it under the same terms as Perl itself.
This makes it easy for people to use your module with Perl. Note that this licensing is neither an endorsement or a requirement, you are of course free to choose any licensing.
If the package has a mailing list, include a URL or subscription instructions here.
If the package has a web site, include a URL here.
In addition, some systems use CONFORMING TO to note conformance to relevant standards and MT-LEVEL to note safeness for use in threaded programs or signal handlers. These headings are primarily useful when documenting parts of a C library. Documentation of object-oriented libraries or modules may use CONSTRUCTORS and METHODS sections for detailed documentation of the parts of the library and save the DESCRIPTION section for an overview; other large modules may use FUNCTIONS for similar reasons. Some people use OVERVIEW to summarize the description if it's quite long.
Section ordering varies, although NAME should always be the first section (you'll break some man page systems otherwise), and NAME, SYNOPSIS, DESCRIPTION, and OPTIONS generally always occur first and in that order if present. In general, SEE ALSO, AUTHOR, and similar material should be left for last. Some systems also move WARNINGS and NOTES to last. The order given above should be reasonable for most purposes.
Finally, as a general note, try not to use an excessive amount of markup. As documented here and in Pod::Man, you can safely leave Perl variables, function names, man page references, and the like unadorned by markup and the POD translators will figure it out for you. This makes it much easier to later edit the documentation. Note that many existing translators (including this one currently) will do the wrong thing with e-mail addresses when wrapped in L<>, so don't do that.
For additional information that may be more accurate for your specific system, see either man(5) or man(7) depending on your system manual section numbering conventions.
The man page documenting the an macro set may be man(5) instead of man(7) on your system.
The current version of this script is always available from its web site at <http://www.eyrie.org/~eagle/software/podlators/>. It is also part of the Perl core distribution as of 5.6.0.
This program is free software; you may redistribute it and/or modify it under the same terms as Perl itself.