Pod::Readme - Convert POD to README file


Pod-Readme documentation Contained in the Pod-Readme distribution.

Index


Code Index:

NAME

Top

Pod::Readme - Convert POD to README file

SYNOPSIS

Top

  use Pod::Readme;
  my $parser = Pod::Readme->new();

  # Read POD from STDIN and write to STDOUT
  $parser->parse_from_filehandle;

  # Read POD from Module.pm and write to README
  $parser->parse_from_file('Module.pm', 'README');

DESCRIPTION

Top

This module is a subclass of Pod::PlainText which provides additional POD markup for generating README files.

Why should one bother with this? One can simply use

  pod2text Module.pm > README

A problem with doing that is that the default pod2text converter will add text to links, so that "L<Module>" is translated to "the Module manpage".

Another problem is that the README includes the entirety of the module documentation! Most people browsing the README file do not need all of this information.

Likewise, including installation and requirement information in the module documentation is not necessary either, since the module is already installed.

This module allows authors to mark portions of the POD to be included only in, or to be excluded from the README file. It also allows you to include portions of another file (such as a separate ChangeLog).

Markup

Special POD markup options are described below:

begin/end
  =begin readme

  =head1 README ONLY

  This section will only show up in the README file.

  =end readme

Delineates a POD section that is only available in README file. If you prefer to include plain text instead, add the text modifier:

  =begin readme text

  README ONLY (PLAINTEXT)

      This section will only show up in the README file.

  =end readme

Note that placing a colon before the section to indicate that it is POD (e.g. begin :readme) is not supported in this version.

stop/continue
  =for readme stop

All POD that follows will not be included in the README, until a continue command occurs:

  =for readme continue

include
  =for readme include file=filename type=type start=Regexp stop=Regexp

  =for readme include file=Changes start=^0.09 stop=^0.081 type=text

Includes a plaintext file named filename, starting with the line that contains the start Regexp and ending at the line that begins with the stop Regexp. (The start and stop Regexps are optional: one or both may be omitted.)

Type may be text or pod. If omitted, pod will be assumed.

Quotes may be used when the filename or marks contains spaces:

  =for readme include file="another file.pod"

One can also using maintain multiple file types (such as including TODO, or COPYING) by using a modified constructor:

  $parser = Pod::Readme->new( readme_type => "copying" );

In the above Markup commands replace "readme" with the tag specified instead (such as "copying"):

  =begin copying

As of version 0.03 you can specify multiple sections by separating them with a comma:

  =begin copying,readme

There is also no standard list of type names. Some names might be recognized by other POD processors (such as "testing" or "html"). Pod::Readme will reject the following "known" type names when they are specified in the constructor:

    testing html xhtml xml docbook rtf man nroff dsr rno latex tex code

You can also use a "debug" mode to diagnose any problems, such as mistyped format names:

  $parser = Pod::Readme->new( debug => 1 );

Warnings will be issued for any ignored formatting commands.

Example

For an example, see the Readme.pm file in this distribution.

SEE ALSO

Top

See perlpod, perlpodspec and podlators.

AUTHOR

Top

Originally by Robert Rothenberg <rrwo at cpan.org>

Now maintained by David Precious <davidp@preshweb.co.uk>

Suggestions, Bug Reporting and Contributing

This module is developed on GitHub at:

http://github.com/bigpresh/Pod-Readme

LICENSE

Top

Copyright (c) 2005,2006 Robert Rothenberg. All rights reserved. This program is free software; you can redistribute it and/or modify it under the same terms as Perl itself.

Some portions are based on Pod::PlainText 2.02.


Pod-Readme documentation Contained in the Pod-Readme distribution.
package Pod::Readme;

use 5.005;
use strict;

use Carp;
use IO::File;
use Pod::PlainText;
use Regexp::Common qw( URI );

use vars qw( @ISA $VERSION );

@ISA = qw( Pod::PlainText );

$VERSION = '0.11';

{
  my %INVALID_TYPES = map { $_ => 1, } (qw(
    test testing tests
    html xhtml xml docbook rtf man nroff dsr rno latex tex code
  ));

  sub initialize {
    my $self = shift;

    $$self{README_SKIP} ||= 0;
    $$self{readme_type} ||= "readme";

    $$self{debug}       ||= 0;

    $self->SUPER::initialize;

    croak "$$self{readme_type} is an invalid readme_type",
      if ($INVALID_TYPES{ $$self{readme_type} });
  }
}

sub output {
  my $self = shift;
  return if $$self{README_SKIP};
  $self->SUPER::output(@_);
}


sub _parse_args {
  my $self   = shift;
  my $string = shift;
  my @values = ( );

  my $arg      = "";
  my $in_quote = 0;
  my $last;
  foreach (split //, $string) {
    if (/\s/) {
      if ($in_quote) {
        $arg .= $_;
      }
      else {
        if ($arg ne "") {
          push @values, $arg;
          $arg = "";
        }
      }
    }
    else {
      $arg .= $_;
      if (/\"/) {
        if ($in_quote) {
          $in_quote = 0 unless ($last eq "\\");
        }
        else {
          # croak "expected \"name=\" before quotes" unless ($last eq "=");
          $in_quote = 1;
        }
      }
    }
    $last = $_;
  }
  push @values, $arg if ($arg ne "");
  return @values;
}



sub cmd_begin {
  my $self = shift;
  my $sec  = $$self{readme_type} || "readme";
  my @fmt  = $self->_parse_args($_[0]);
  my %secs = map { $_ => 1, } split /,/, $fmt[0];
  if ($secs{$sec}) {
    $$self{README_SKIP} = 0;
    if (($fmt[1]||"pod") eq "pod") {
    }
    elsif ($fmt[1] eq "text") {
      $$self{VERBATIM} = 1;
    }
    else {
      # TODO - return error
      $$self{EXCLUDE}  = 1;
    }
  }
  else {
    carp "Ignoring document type(s) \"$fmt[0]\" in POD line $_[1]"
      if ($$self{debug});
    $self->SUPER::cmd_begin(@_);
  }
}


sub cmd_for {
  my $self = shift;
  my $sec  = $$self{readme_type} || "readme";
  my @fmt  = $self->_parse_args($_[0]);
  my %secs = map { $_ => 1, } split /,/, $fmt[0];
  if ($secs{$sec}) {
    my $cmd = $fmt[1] || "continue";
    if ($cmd eq "stop") {
      $$self{README_SKIP} = 1;
    } elsif ($cmd eq "continue") {
      $$self{README_SKIP} = 0;
    } elsif ($cmd eq "include") {

      my %arg = map {
        s/\"//g;
        my ($k,$v) = split /\=/;
        $k => $v;
      } @fmt[2..$#fmt];
      $arg{type} ||= "pod";

      my $text =
	$self->_include_file( map { $arg{$_} } (qw( type file start stop )) );
      if ($arg{type} eq "text") {
        $self->verbatim($text, $_[1], $_[2]);
      } else {
        $self->textblock($text, $_[1], $_[2]);
      }
    } else {
      croak "Don\'t know how to \"$cmd\" in POD line $_[1]";
    }
  }
  else {
    carp "Ignoring document type(s) \"$fmt[0]\" in POD line $_[1]"
      if ($$self{debug});
    $self->SUPER::cmd_for(@_);
  }
}

sub cmd_encoding {
    my $self = shift;
    my $encoding = (split /\s+/, shift)[0];
    if ($self->{_encoding}) {
        die "=encoding option must occur only once!";
    }
    $self->{_encoding} = $encoding;
    # TODO: Need to actually do something with this option
    # At least recognising it and not dying is a step in the right direction.
}

sub _include_file {
  my $self = shift;
  my $type = shift || "pod";
  my $file = shift;
  my $mark = shift || "";
  my $stop = shift || "";

  my $fh   = IO::File->new("<$file")
    || croak "Unable to open file \"$file\"";

  my $buffer = "";
  while (my $line = <$fh>) {
    next if (($mark ne "") && ($line !~ /$mark/));
    $mark = "" if ($mark ne "");
    last if (($stop ne "") && ($line =~ /$stop/));
    $buffer .= $line;
  }
  close $fh;

  if ($type ne "pod") {
    my $indent = " " x $$self{MARGIN};
    $buffer =~ s/([\r\n]+)(\t)?/$1 . $indent x (1+length($2||""))/ge;
    $buffer =~ s/($indent)+$//;
  }

  return $buffer;
}


# This code is based on code from Pod::PlainText 2.02

sub seq_l {
  my $self = shift;
  local $_ = shift;
    # Smash whitespace in case we were split across multiple lines.
    s/\s+/ /g;

    # If we were given any explicit text, just output it.
    if (/^([^|]+)\|/) { return $1 }

    # Okay, leading and trailing whitespace isn't important; get rid of it.
    s/^\s+//;
    s/\s+$//;

    # Default to using the whole content of the link entry as a section
    # name.  Note that L<manpage/> forces a manpage interpretation, as does
    # something looking like L<manpage(section)>.  The latter is an
    # enhancement over the original Pod::Text.


    my ($manpage, $section) = ('', $_);
    if (/$RE{URI}/ || /^(?:https?|ftps?|svn):/) {
        # a URL
        return $_;
    } elsif (/^"\s*(.*?)\s*"$/) {
        $section = '"' . $1 . '"';
    } elsif (m/^[-:.\w]+(?:\(\S+\))?$/) {
        ($manpage, $section) = ($_, '');
    } elsif (m%/%) {
        ($manpage, $section) = split (/\s*\/\s*/, $_, 2);
    }

    if (length $manpage) {
      return $manpage;
    } else {
      return $section;
    }
}