text.pm

MSYS在windows下模拟了一个类unix的终端

# Pod::Text -- Convert POD data to formatted ASCII text.
# $Id: Text.pm,v 2.8 2001/02/10 06:50:23 eagle Exp $
#
# Copyright 1999, 2000, 2001 by Russ Allbery <[e-mail]>
#
# This program is free software; you can redistribute it and/or modify it
# under the same terms as Perl itself.
#
# This module is intended to be a replacement for Pod::Text, and attempts to
# match its output except for some specific circumstances where other
# decisions seemed to produce better output.  It uses Pod::Parser and is
# designed to be very easy to subclass.

############################################################################
# Modules and declarations
############################################################################

package Pod::Text;

require 5.004;

use Carp qw(carp croak);
use Exporter ();
use Pod::Select ();

use strict;
use vars qw(@ISA @EXPORT %ESCAPES $VERSION);

# We inherit from Pod::Select instead of Pod::Parser so that we can be used
# by Pod::Usage.
@ISA = qw(Pod::Select Exporter);

# We have to export pod2text for backward compatibility.
@EXPORT = qw(pod2text);

# Don't use the CVS revision as the version, since this module is also in
# Perl core and too many things could munge CVS magic revision strings.
# This number should ideally be the same as the CVS revision in podlators,
# however.
$VERSION = 2.08;


############################################################################
# Table of supported E<> escapes
############################################################################

# This table is taken near verbatim from Pod::PlainText in Pod::Parser,
# which got it near verbatim from the original Pod::Text.  It is therefore
# credited to Tom Christiansen, and I'm glad I didn't have to write it.  :)
# "iexcl" to "divide" added by Tim Jenness.
%ESCAPES = (
    'amp'       =>    '&',      # ampersand
    'lt'        =>    '<',      # left chevron, less-than
    'gt'        =>    '>',      # right chevron, greater-than
    'quot'      =>    '"',      # double quote
    'sol'       =>    '/',      # solidus (forward slash)
    'verbar'    =>    '|',      # vertical bar

    "Aacute"    =>    "\xC1",   # capital A, acute accent
    "aacute"    =>    "\xE1",   # small a, acute accent
    "Acirc"     =>    "\xC2",   # capital A, circumflex accent
    "acirc"     =>    "\xE2",   # small a, circumflex accent
    "AElig"     =>    "\xC6",   # capital AE diphthong (ligature)
    "aelig"     =>    "\xE6",   # small ae diphthong (ligature)
    "Agrave"    =>    "\xC0",   # capital A, grave accent
    "agrave"    =>    "\xE0",   # small a, grave accent
    "Aring"     =>    "\xC5",   # capital A, ring
    "aring"     =>    "\xE5",   # small a, ring
    "Atilde"    =>    "\xC3",   # capital A, tilde
    "atilde"    =>    "\xE3",   # small a, tilde
    "Auml"      =>    "\xC4",   # capital A, dieresis or umlaut mark
    "auml"      =>    "\xE4",   # small a, dieresis or umlaut mark
    "Ccedil"    =>    "\xC7",   # capital C, cedilla
    "ccedil"    =>    "\xE7",   # small c, cedilla
    "Eacute"    =>    "\xC9",   # capital E, acute accent
    "eacute"    =>    "\xE9",   # small e, acute accent
    "Ecirc"     =>    "\xCA",   # capital E, circumflex accent
    "ecirc"     =>    "\xEA",   # small e, circumflex accent
    "Egrave"    =>    "\xC8",   # capital E, grave accent
    "egrave"    =>    "\xE8",   # small e, grave accent
    "ETH"       =>    "\xD0",   # capital Eth, Icelandic
    "eth"       =>    "\xF0",   # small eth, Icelandic
    "Euml"      =>    "\xCB",   # capital E, dieresis or umlaut mark
    "euml"      =>    "\xEB",   # small e, dieresis or umlaut mark
    "Iacute"    =>    "\xCC",   # capital I, acute accent
    "iacute"    =>    "\xEC",   # small i, acute accent
    "Icirc"     =>    "\xCE",   # capital I, circumflex accent
    "icirc"     =>    "\xEE",   # small i, circumflex accent
    "Igrave"    =>    "\xCD",   # capital I, grave accent
    "igrave"    =>    "\xED",   # small i, grave accent
    "Iuml"      =>    "\xCF",   # capital I, dieresis or umlaut mark
    "iuml"      =>    "\xEF",   # small i, dieresis or umlaut mark
    "Ntilde"    =>    "\xD1",   # capital N, tilde
    "ntilde"    =>    "\xF1",   # small n, tilde
    "Oacute"    =>    "\xD3",   # capital O, acute accent
    "oacute"    =>    "\xF3",   # small o, acute accent
    "Ocirc"     =>    "\xD4",   # capital O, circumflex accent
    "ocirc"     =>    "\xF4",   # small o, circumflex accent
    "Ograve"    =>    "\xD2",   # capital O, grave accent
    "ograve"    =>    "\xF2",   # small o, grave accent
    "Oslash"    =>    "\xD8",   # capital O, slash
    "oslash"    =>    "\xF8",   # small o, slash
    "Otilde"    =>    "\xD5",   # capital O, tilde
    "otilde"    =>    "\xF5",   # small o, tilde
    "Ouml"      =>    "\xD6",   # capital O, dieresis or umlaut mark
    "ouml"      =>    "\xF6",   # small o, dieresis or umlaut mark
    "szlig"     =>    "\xDF",   # small sharp s, German (sz ligature)
    "THORN"     =>    "\xDE",   # capital THORN, Icelandic
    "thorn"     =>    "\xFE",   # small thorn, Icelandic
    "Uacute"    =>    "\xDA",   # capital U, acute accent
    "uacute"    =>    "\xFA",   # small u, acute accent
    "Ucirc"     =>    "\xDB",   # capital U, circumflex accent
    "ucirc"     =>    "\xFB",   # small u, circumflex accent
    "Ugrave"    =>    "\xD9",   # capital U, grave accent
    "ugrave"    =>    "\xF9",   # small u, grave accent
    "Uuml"      =>    "\xDC",   # capital U, dieresis or umlaut mark
    "uuml"      =>    "\xFC",   # small u, dieresis or umlaut mark
    "Yacute"    =>    "\xDD",   # capital Y, acute accent
    "yacute"    =>    "\xFD",   # small y, acute accent
    "yuml"      =>    "\xFF",   # small y, dieresis or umlaut mark

    "laquo"     =>    "\xAB",   # left pointing double angle quotation mark
    "lchevron"  =>    "\xAB",   #  synonym (backwards compatibility)
    "raquo"     =>    "\xBB",   # right pointing double angle quotation mark
    "rchevron"  =>    "\xBB",   #  synonym (backwards compatibility)

    "iexcl"     =>    "\xA1",   # inverted exclamation mark
    "cent"      =>    "\xA2",   # cent sign
    "pound"     =>    "\xA3",   # (UK) pound sign
    "curren"    =>    "\xA4",   # currency sign
    "yen"       =>    "\xA5",   # yen sign
    "brvbar"    =>    "\xA6",   # broken vertical bar
    "sect"      =>    "\xA7",   # section sign
    "uml"       =>    "\xA8",   # diaresis
    "copy"      =>    "\xA9",   # Copyright symbol
    "ordf"      =>    "\xAA",   # feminine ordinal indicator
    "not"       =>    "\xAC",   # not sign
    "shy"       =>    "\xAD",   # soft hyphen
    "reg"       =>    "\xAE",   # registered trademark
    "macr"      =>    "\xAF",   # macron, overline
    "deg"       =>    "\xB0",   # degree sign
    "plusmn"    =>    "\xB1",   # plus-minus sign
    "sup2"      =>    "\xB2",   # superscript 2
    "sup3"      =>    "\xB3",   # superscript 3
    "acute"     =>    "\xB4",   # acute accent
    "micro"     =>    "\xB5",   # micro sign
    "para"      =>    "\xB6",   # pilcrow sign = paragraph sign
    "middot"    =>    "\xB7",   # middle dot = Georgian comma
    "cedil"     =>    "\xB8",   # cedilla
    "sup1"      =>    "\xB9",   # superscript 1
    "ordm"      =>    "\xBA",   # masculine ordinal indicator
    "frac14"    =>    "\xBC",   # vulgar fraction one quarter
    "frac12"    =>    "\xBD",   # vulgar fraction one half
    "frac34"    =>    "\xBE",   # vulgar fraction three quarters
    "iquest"    =>    "\xBF",   # inverted question mark
    "times"     =>    "\xD7",   # multiplication sign
    "divide"    =>    "\xF7",   # division sign
);


############################################################################
# Initialization
############################################################################

# Initialize the object.  Must be sure to call our parent initializer.
sub initialize {
    my $self = shift;

    $$self{alt}      = 0  unless defined $$self{alt};
    $$self{indent}   = 4  unless defined $$self{indent};
    $$self{loose}    = 0  unless defined $$self{loose};
    $$self{sentence} = 0  unless defined $$self{sentence};
    $$self{width}    = 76 unless defined $$self{width};

    # Figure out what quotes we'll be using for C<> text.
    $$self{quotes} ||= '"';
    if ($$self{quotes} eq 'none') {
        $$self{LQUOTE} = $$self{RQUOTE} = '';
    } elsif (length ($$self{quotes}) == 1) {
        $$self{LQUOTE} = $$self{RQUOTE} = $$self{quotes};
    } elsif ($$self{quotes} =~ /^(.)(.)$/
             || $$self{quotes} =~ /^(..)(..)$/) {
        $$self{LQUOTE} = $1;
        $$self{RQUOTE} = $2;
    } else {
        croak qq(Invalid quote specification "$$self{quotes}");
    }

    $$self{INDENTS}  = [];              # Stack of indentations.
    $$self{MARGIN}   = $$self{indent};  # Current left margin in spaces.

    $self->SUPER::initialize;
}


############################################################################
# Core overrides
############################################################################

# Called for each command paragraph.  Gets the command, the associated
# paragraph, the line number, and a Pod::Paragraph object.  Just dispatches
# the command to a method named the same as the command.  =cut is handled
# internally by Pod::Parser.
sub command {
    my $self = shift;
    my $command = shift;
    return if $command eq 'pod';
    return if ($$self{EXCLUDE} && $command ne 'end');
    $self->item ("\n") if defined $$self{ITEM};
    if ($self->can ('cmd_' . $command)) {
        $command = 'cmd_' . $command;
        $self->$command (@_);
    } else {
        my ($text, $line, $paragraph) = @_;
        my $file;
        ($file, $line) = $paragraph->file_line;
        $text =~ s/\n+\z//;
        $text = " $text" if ($text =~ /^\S/);
        warn qq($file:$line: Unknown command paragraph "=$command$text"\n);
        return;
    }
}

# Called for a verbatim paragraph.  Gets the paragraph, the line number, and
# a Pod::Paragraph object.  Just output it verbatim, but with tabs converted
# to spaces.
sub verbatim {
    my $self = shift;
    return if $$self{EXCLUDE};
    $self->item if defined $$self{ITEM};
    local $_ = shift;
    return if /^\s*$/;
    s/^(\s*\S+)/(' ' x $$self{MARGIN}) . $1/gme;
    $self->output ($_);
}

# Called for a regular text block.  Gets the paragraph, the line number, and
# a Pod::Paragraph object.  Perform interpolation and output the results.
sub textblock {
    my $self = shift;
    return if $$self{EXCLUDE};
    $self->output ($_[0]), return if $$self{VERBATIM};
    local $_ = shift;
    my $line = shift;

    # Perform a little magic to collapse multiple L<> references.  This is
    # here mostly for backwards-compatibility.  We'll just rewrite the whole
    # thing into actual text at this part, bypassing the whole internal
    # sequence parsing thing.
    s{
        (
          L<                    # A link of the form L</something>.
              /
              (
                  [:\w]+        # The item has to be a simple word...
                  (\(\))?       # ...or simple function.
              )
          >
          (
              ,?\s+(and\s+)?    # Allow lots of them, conjuncted.
              L<
                  /
                  (
                      [:\w]+
                      (\(\))?
                  )
              >
          )+
        )
    } {
        local $_ = $1;
        s%L</([^>]+)>%$1%g;
        my @items = split /(?:,?\s+(?:and\s+)?)/;
        my $string = "the ";
        my $i;
        for ($i = 0; $i < @items; $i++) {
            $string .= $items[$i];
            $string .= ", " if @items > 2 && $i != $#items;
            $string .= " and " if ($i == $#items - 1);
        }
        $string .= " entries elsewhere in this document";
        $string;
    }gex;

    # Now actually interpolate and output the paragraph.
    $_ = $self->interpolate ($_, $line);
    s/\s+$/\n/;
    if (defined $$self{ITEM}) {
        $self->item ($_ . "\n");
    } else {
        $self->output ($self->reformat ($_ . "\n"));
    }
}

# Called for an interior sequence.  Gets the command, argument, and a
# Pod::InteriorSequence object and is expected to return the resulting text.
# Calls code, bold, italic, file, and link to handle those types of
# sequences, and handles S<>, E<>, X<>, and Z<> directly.
sub interior_sequence {
    my $self = shift;
    my $command = shift;
    local $_ = shift;
    return '' if ($command eq 'X' || $command eq 'Z');

    # Expand escapes into the actual character now, carping if invalid.
    if ($command eq 'E') {
        if (/^\d+$/) {
            return chr;
        } else {
            return $ESCAPES{$_} if defined $ESCAPES{$_};
            carp "Unknown escape: E<$_>";
            return "E<$_>";
        }
    }

    # For all the other sequences, empty content produces no output.
    return if $_ eq '';

    # For S<>, compress all internal whitespace and then map spaces to \01.
    # When we output the text, we'll map this back.
    if ($command eq 'S') {
        s/\s{2,}/ /g;
        tr/ /\01/;
        return $_;
    }

    # Anything else needs to get dispatched to another method.
    if    ($command eq 'B') { return $self->seq_b ($_) }
    elsif ($command eq 'C') { return $self->seq_c ($_) }
    elsif ($command eq 'F') { return $self->seq_f ($_) }
    elsif ($command eq 'I') { return $self->seq_i ($_) }
    elsif ($command eq 'L') { return $self->seq_l ($_) }
    else { carp "Unknown sequence $command<$_>" }
}

# Called for each paragraph that's actually part of the POD.  We take
# advantage of this opportunity to untabify the input.
sub preprocess_paragraph {
    my $self = shift;
    local $_ = shift;
    1 while s/^(.*?)(\t+)/$1 . ' ' x (length ($2) * 8 - length ($1) % 8)/me;
    $_;
}


############################################################################
# Command paragraphs
############################################################################

# All command paragraphs take the paragraph and the line number.

# First level heading.
sub cmd_head1 {
    my $self = shift;
    local $_ = shift;
    s/\s+$//;
    $_ = $self->interpolate ($_, shift);
    if ($$self{alt}) {
        $self->output ("\n==== $_ ====\n\n");
    } else {
        $_ .= "\n" if $$self{loose};
        $self->output ($_ . "\n");
    }
}

# Second level heading.
sub cmd_head2 {
    my $self = shift;
    local $_ = shift;
    s/\s+$//;
    $_ = $self->interpolate ($_, shift);
    if ($$self{alt}) {
        $self->output ("\n==   $_   ==\n\n");
    } else {
        $self->output (' ' x ($$self{indent} / 2) . $_ . "\n\n");
    }
}

# Third level heading.
sub cmd_head3 {
    my $self = shift;
    local $_ = shift;
    s/\s+$//;
    $_ = $self->interpolate ($_, shift);
    if ($$self{alt}) {
        $self->output ("\n=    $_    =\n\n");
    } else {
        $self->output (' ' x ($$self{indent} * 2 / 3 + 0.5) . $_ . "\n\n");
    }
}

# Third level heading.
sub cmd_head4 {
    my $self = shift;
    local $_ = shift;
    s/\s+$//;
    $_ = $self->interpolate ($_, shift);
    if ($$self{alt}) {
        $self->output ("\n-    $_    -\n\n");
    } else {
        $self->output (' ' x ($$self{indent} * 3 / 4 + 0.5) . $_ . "\n\n");
    }
}

# Start a list.
sub cmd_over {
    my $self = shift;
    local $_ = shift;
    unless (/^[-+]?\d+\s+$/) { $_ = $$self{indent} }
    push (@{ $$self{INDENTS} }, $$self{MARGIN});
    $$self{MARGIN} += ($_ + 0);
}

# End a list.