# Pod::Text -- Convert POD data to formatted ASCII text.
-# $Id: Text.pm,v 2.13 2001/10/20 08:07:21 eagle Exp $
+# $Id: Text.pm,v 2.21 2002/08/04 03:34:58 eagle Exp $
#
-# Copyright 1999, 2000, 2001 by Russ Allbery <rra@stanford.edu>
+# Copyright 1999, 2000, 2001, 2002 by Russ Allbery <rra@stanford.edu>
#
# This program is free software; you may redistribute it and/or modify it
# under the same terms as Perl itself.
#
-# This module replaces the old Pod::Text that came with versions of Perl prior
-# to 5.6.0, 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.
+# This module converts POD to formatted text. It replaces the old Pod::Text
+# module that came with versions of Perl prior to 5.6.0 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.
#
# Perl core hackers, please note that this module is also separately
# maintained outside of the Perl core as part of the podlators. Please send
use Carp qw(carp croak);
use Exporter ();
+use Pod::ParseLink qw(parselink);
use Pod::Select ();
use strict;
# 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.
-$VERSION = 2.13;
+$VERSION = 2.21;
##############################################################################
# "divide" added by Tim Jenness.
%ESCAPES = (
'amp' => '&', # ampersand
+ 'apos' => "'", # apostrophe
'lt' => '<', # left chevron, less-than
'gt' => '>', # right chevron, greater-than
'quot' => '"', # double quote
"copy" => "\xA9", # Copyright symbol
"ordf" => "\xAA", # feminine ordinal indicator
"not" => "\xAC", # not sign
- "shy" => "\xAD", # soft hyphen
+ "shy" => '', # soft (discretionary) hyphen
"reg" => "\xAE", # registered trademark
"macr" => "\xAF", # macron, overline
"deg" => "\xB0", # degree sign
"iquest" => "\xBF", # inverted question mark
"times" => "\xD7", # multiplication sign
"divide" => "\xF7", # division sign
+
+ "nbsp" => "\x01", # non-breaking space
);
$$self{alt} = 0 unless defined $$self{alt};
$$self{indent} = 4 unless defined $$self{indent};
+ $$self{margin} = 0 unless defined $$self{margin};
$$self{loose} = 0 unless defined $$self{loose};
$$self{sentence} = 0 unless defined $$self{sentence};
$$self{width} = 76 unless defined $$self{width};
croak qq(Invalid quote specification "$$self{quotes}");
}
- $$self{INDENTS} = []; # Stack of indentations.
- $$self{MARGIN} = $$self{indent}; # Current left margin in spaces.
+ # Stack of indentations.
+ $$self{INDENTS} = [];
+
+ # Current left margin.
+ $$self{MARGIN} = $$self{indent} + $$self{margin};
$self->SUPER::initialize;
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 (@_);
($file, $line) = $paragraph->file_line;
$text =~ s/\n+\z//;
$text = " $text" if ($text =~ /^\S/);
- warn qq($file:$line: Unknown command paragraph "=$command$text"\n);
+ warn qq($file:$line: Unknown command paragraph: =$command$text\n);
return;
}
}
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.
+ # Interpolate and output the paragraph.
$_ = $self->interpolate ($_, $line);
s/\s+$/\n/;
if (defined $$self{ITEM}) {
}
}
-# Called for an interior sequence. Gets the command, argument, and a
+# Called for a formatting code. 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.
+# Calls methods for code, bold, italic, file, and link to handle those types
+# of codes, and handles S<>, E<>, X<>, and Z<> directly.
sub interior_sequence {
- my $self = shift;
- my $command = shift;
- local $_ = shift;
+ local $_;
+ my ($self, $command, $seq);
+ ($self, $command, $_, $seq) = @_;
+
+ # We have to defer processing of the inside of an L<> formatting code. If
+ # this code is nested inside an L<> code, return the literal raw text of
+ # it.
+ my $parent = $seq->nested;
+ while (defined $parent) {
+ return $seq->raw_text if ($parent->cmd_name eq 'L');
+ $parent = $parent->nested;
+ }
+
+ # Index entries are ignored in plain text.
return '' if ($command eq 'X' || $command eq 'Z');
# Expand escapes into the actual character now, warning if invalid.
return chr;
} else {
return $ESCAPES{$_} if defined $ESCAPES{$_};
- my $seq = shift;
my ($file, $line) = $seq->file_line;
warn "$file:$line: Unknown escape: E<$_>\n";
return "E<$_>";
}
}
- # For all the other sequences, empty content produces no output.
+ # For all the other formatting codes, 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;
+ s/\s+/ /g;
tr/ /\01/;
return $_;
}
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 ($_) }
+ elsif ($command eq 'L') { return $self->seq_l ($_, $seq) }
else {
- my $seq = shift;
my ($file, $line) = $seq->file_line;
- warn "$file:$line: Unknown sequence $command<$_>\n";
+ warn "$file:$line: Unknown formatting code: $command<$_>\n";
}
}
# 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 $$self{loose};
- $self->output ($_ . "\n");
- }
+ my ($self, $text, $line) = @_;
+ $self->heading ($text, $line, 0, '====');
}
# Second level heading.
sub cmd_head2 {
- my $self = shift;
- local $_ = shift;
- s/\s+$//;
- $_ = $self->interpolate ($_, shift);
- if ($$self{alt}) {
- $self->output ("\n== $_ ==\n\n");
- } else {
- $self->output (' ' x ($$self{indent} / 2) . $_ . "\n\n");
- }
+ my ($self, $text, $line) = @_;
+ $self->heading ($text, $line, $$self{indent} / 2, '== ');
}
# 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) . $_ . "\n\n");
- }
+ my ($self, $text, $line) = @_;
+ $self->heading ($text, $line, $$self{indent} * 2 / 3 + 0.5, '= ');
}
# Third level heading.
sub cmd_head4 {
- my $self = shift;
- local $_ = shift;
- s/\s+$//;
- $_ = $self->interpolate ($_, shift);
- if ($$self{alt}) {
- $self->output ("\n- $_ -\n\n");
- } else {
- $self->output (' ' x ($$self{indent} * 3 / 4 + 0.5) . $_ . "\n\n");
- }
+ my ($self, $text, $line) = @_;
+ $self->heading ($text, $line, $$self{indent} * 3 / 4 + 0.5, '- ');
}
# Start a list.
sub cmd_over {
my $self = shift;
local $_ = shift;
+ $self->item ("\n\n") if defined $$self{ITEM};
unless (/^[-+]?\d+\s+$/) { $_ = $$self{indent} }
push (@{ $$self{INDENTS} }, $$self{MARGIN});
$$self{MARGIN} += ($_ + 0);
# End a list.
sub cmd_back {
my ($self, $text, $line, $paragraph) = @_;
+ $self->item ("\n\n") if defined $$self{ITEM};
$$self{MARGIN} = pop @{ $$self{INDENTS} };
unless (defined $$self{MARGIN}) {
my $file;
if (defined $$self{ITEM}) { $self->item }
local $_ = shift;
s/\s+$//;
- $$self{ITEM} = $self->interpolate ($_);
+ $$self{ITEM} = $_ ? $self->interpolate ($_) : '*';
}
# Begin a block for a particular translator. Setting VERBATIM triggers
##############################################################################
-# Interior sequences
+# Formatting codes
##############################################################################
-# The simple formatting ones. These are here mostly so that subclasses can
-# override them and do more complicated things.
+# The simple ones. These are here mostly so that subclasses can override them
+# and do more complicated things.
sub seq_b { return $_[0]{alt} ? "``$_[1]''" : $_[1] }
sub seq_f { return $_[0]{alt} ? "\"$_[1]\"" : $_[1] }
sub seq_i { return '*' . $_[1] . '*' }
| \$+ [\#^]? \S $index # special ($^Foo, $")
| [\$\@%&*]+ \#? [:\'\w]+ $index # plain var or func
| [\$\@%&*]* [:\'\w]+ (?: -> )? \(\s*[^\s,]\s*\) # 0/1-arg func call
- | [+-]? [\d.]+ (?: [eE] [+-]? \d+ )? # a number
+ | [+-]? ( \d[\d.]* | \.\d+ ) (?: [eE][+-]?\d+ )? # a number
| 0x [a-fA-F\d]+ # a hex constant
)
\s*\z
return $$self{alt} ? "``$_''" : "$$self{LQUOTE}$_$$self{RQUOTE}";
}
-# The complicated one. Handle links. Since this is plain text, we can't
-# actually make any real links, so this is all to figure out what text we
-# print out.
+# Handle links. Since this is plain text, we can't actually make any real
+# links, so this is all to figure out what text we print out. Most of the
+# work is done by Pod::ParseLink.
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 $1 }
+ my ($self, $link, $seq) = @_;
+ my ($text, $type) = (parselink ($link))[1,4];
+ my ($file, $line) = $seq->file_line;
+ $text = $self->interpolate ($text, $line);
+ $text = '<' . $text . '>' if $type eq 'url';
+ return $text || '';
+}
- # Okay, leading and trailing whitespace isn't important; get rid of it.
- s/^\s+//;
- s/\s+$//;
- # If the argument looks like a URL, return it verbatim. This only handles
- # URLs that use the server syntax.
- if (m%^[a-z]+://\S+$%) { return $_ }
-
- # Default to using the whole content of the link entry as a section name.
- # Note that L<manpage/> forces a manpage interpretation, as does something
- # looking like L<manpage(section)>. The latter is an enhancement over the
- # original Pod::Text.
- my ($manpage, $section) = ('', $_);
- if (/^"\s*(.*?)\s*"$/) {
- $section = '"' . $1 . '"';
- } elsif (m/^[-:.\w]+(?:\(\S+\))?$/) {
- ($manpage, $section) = ($_, '');
- } elsif (m%/%) {
- ($manpage, $section) = split (/\s*\/\s*/, $_, 2);
- }
+##############################################################################
+# Header handling
+##############################################################################
- # Now build the actual output text.
- my $text = '';
- if (!length $section) {
- $text = "the $manpage manpage" if length $manpage;
- } elsif ($section =~ /^[:\w]+(?:\(\))?/) {
- $text .= 'the ' . $section . ' entry';
- $text .= (length $manpage) ? " in the $manpage manpage"
- : " elsewhere in this document";
+# The common code for handling all headers. Takes the interpolated header
+# text, the line number, the indentation, and the surrounding marker for the
+# alt formatting method.
+sub heading {
+ my ($self, $text, $line, $indent, $marker) = @_;
+ $self->item ("\n\n") if defined $$self{ITEM};
+ $text =~ s/\s+$//;
+ $text = $self->interpolate ($text, $line);
+ if ($$self{alt}) {
+ my $closemark = reverse (split (//, $marker));
+ my $margin = ' ' x $$self{margin};
+ $self->output ("\n" . "$margin$marker $text $closemark" . "\n\n");
} else {
- $section =~ s/^\"\s*//;
- $section =~ s/\s*\"$//;
- $text .= 'the section on "' . $section . '"';
- $text .= " in the $manpage manpage" if length $manpage;
+ $text .= "\n" if $$self{loose};
+ my $margin = ' ' x ($$self{margin} + $indent);
+ $self->output ($margin . $text . "\n");
}
- $text;
}
undef $$self{ITEM};
my $indent = $$self{INDENTS}[-1];
unless (defined $indent) { $indent = $$self{indent} }
- my $space = ' ' x $indent;
- $space =~ s/^ /:/ if $$self{alt};
+ my $margin = ' ' x $$self{margin};
if (!$_ || /^\s+$/ || ($$self{MARGIN} - $indent < length ($tag) + 1)) {
- my $margin = $$self{MARGIN};
+ my $realindent = $$self{MARGIN};
$$self{MARGIN} = $indent;
my $output = $self->reformat ($tag);
+ $output =~ s/^$margin /$margin:/ if ($$self{alt} && $indent > 0);
$output =~ s/\n*$/\n/;
+
+ # If the text is just whitespace, we have an empty item paragraph;
+ # this can result from =over/=item/=back without any intermixed
+ # paragraphs. Insert some whitespace to keep the =item from merging
+ # into the next paragraph.
+ $output .= "\n" if $_ && $_ =~ /^\s*$/;
+
$self->output ($output);
- $$self{MARGIN} = $margin;
- $self->output ($self->reformat ($_)) if /\S/;
+ $$self{MARGIN} = $realindent;
+ $self->output ($self->reformat ($_)) if $_ && /\S/;
} else {
+ my $space = ' ' x $indent;
+ $space =~ s/^$margin /$margin:/ if $$self{alt};
$_ = $self->reformat ($_);
- s/^ /:/ if ($$self{alt} && $indent > 0);
+ s/^$margin /$margin:/ if ($$self{alt} && $indent > 0);
my $tagspace = ' ' x length $tag;
s/^($space)$tagspace/$1$tag/ or warn "Bizarre space in item";
$self->output ($_);
As a derived class from Pod::Parser, Pod::Text supports the same methods and
interfaces. See L<Pod::Parser> for all the details; briefly, one creates a
-new parser with C<Pod::Text-E<gt>new()> and then calls either
+new parser with C<< Pod::Text->new() >> and then calls either
parse_from_filehandle() or parse_from_file().
new() can take options, in the form of key/value pairs, that control the
arbitrary text documents, setting this to true may result in more pleasing
output.
+=item margin
+
+The width of the left margin in spaces. Defaults to 0. This is the margin
+for all text, including headings, not the amount by which regular text is
+indented; for the latter, see the I<indent> option. To set the right
+margin, see the I<width> option.
+
=item quotes
Sets the quote marks used to surround CE<lt>> text. If the value is a
(F) The quote specification given (the quotes option to the constructor) was
invalid. A quote specification must be one, two, or four characters long.
-=item %s:%d: Unknown command paragraph "%s".
+=item %s:%d: Unknown command paragraph: %s
(W) The POD source contained a non-standard command paragraph (something of
the form C<=command args>) that Pod::Man didn't know about. It was ignored.
(W) The POD source contained an C<EE<lt>E<gt>> escape that Pod::Text didn't
know about.
-=item %s:%d: Unknown sequence: %s
+=item %s:%d: Unknown formatting code: %s
-(W) The POD source contained a non-standard internal sequence (something of
+(W) The POD source contained a non-standard formatting code (something of
the form C<XE<lt>E<gt>>) that Pod::Text didn't know about.
=item %s:%d: Unmatched =back
The original Pod::Text contained code to do formatting via termcap
sequences, although it wasn't turned on by default and it was problematic to
get it to work at all. This rewrite doesn't even try to do that, but a
-subclass of it does. Look for L<Pod::Text::Termcap|Pod::Text::Termcap>.
+subclass of it does. Look for L<Pod::Text::Termcap>.
=head1 SEE ALSO
-L<Pod::Parser|Pod::Parser>, L<Pod::Text::Termcap|Pod::Text::Termcap>,
-pod2text(1)
+L<Pod::Parser>, L<Pod::Text::Termcap>, L<pod2text(1)>
+
+The current version of this module is always available from its web site at
+L<http://www.eyrie.org/~eagle/software/podlators/>. It is also part of the
+Perl core distribution as of 5.6.0.
=head1 AUTHOR
-Russ Allbery E<lt>rra@stanford.eduE<gt>, based I<very> heavily on the
-original Pod::Text by Tom Christiansen E<lt>tchrist@mox.perl.comE<gt> and
-its conversion to Pod::Parser by Brad Appleton
-E<lt>bradapp@enteract.comE<gt>.
+Russ Allbery <rra@stanford.edu>, based I<very> heavily on the original
+Pod::Text by Tom Christiansen <tchrist@mox.perl.com> and its conversion to
+Pod::Parser by Brad Appleton <bradapp@enteract.com>.
=head1 COPYRIGHT AND LICENSE
-Copyright 1999, 2000, 2001 by Russ Allbery <rra@stanford.edu>.
+Copyright 1999, 2000, 2001, 2002 by Russ Allbery <rra@stanford.edu>.
This program is free software; you may redistribute it and/or modify it
under the same terms as Perl itself.
projects / p5sagit/p5-mst-13.2.git / blobdiff