<h2>Editing File: /lib64/perl5/B/Terse.pm</h2><form method='post'><textarea name='content' class='file-edit'>package B::Terse;

our $VERSION = '1.07';

use strict;
use B qw(class @specialsv_name);
use B::Concise qw(concise_subref set_style_standard);
use Carp;

sub terse {
    my ($order, $subref) = @_;
    set_style_standard(&quot;terse&quot;);
    if ($order eq &quot;exec&quot;) {
	concise_subref('exec', $subref);
    } else {
	concise_subref('basic', $subref);
    }
}

sub compile {
    my @args = @_;
    my $order = @args ? shift(@args) : &quot;&quot;;
    $order = &quot;-exec&quot; if $order eq &quot;exec&quot;;
    unshift @args, $order if $order ne &quot;&quot;;
    B::Concise::compile(&quot;-terse&quot;, @args);
}

sub indent {
    my ($level) = @_ ? shift : 0;
    return &quot;    &quot; x $level;
}

# Don't use this, at least on OPs in subroutines: it has no way of
# getting to the pad, and will give wrong answers or crash.
sub B::OP::terse {
    carp &quot;B::OP::terse is deprecated and will go away in Perl 5.28; use B::Concise instead&quot;;
    B::Concise::b_terse(@_);
}

sub B::SV::terse {
    my($sv, $level) = (@_, 0);
    my %info;
    B::Concise::concise_sv($sv, \%info);
    my $s = indent($level)
	. B::Concise::fmt_line(\%info, $sv,
				 &quot;#svclass~(?((#svaddr))?)~#svval&quot;, 0);
    chomp $s;
    print &quot;$s\n&quot; unless defined wantarray;
    $s;
}

sub B::NULL::terse {
    my ($sv, $level) = (@_, 0);
    my $s = indent($level) . sprintf &quot;%s (0x%lx)&quot;, class($sv), $$sv;
    print &quot;$s\n&quot; unless defined wantarray;
    $s;
}

sub B::SPECIAL::terse {
    my ($sv, $level) = (@_, 0);
    my $s = indent($level)
	. sprintf( &quot;%s #%d %s&quot;, class($sv), $$sv, $specialsv_name[$$sv]);
    print &quot;$s\n&quot; unless defined wantarray;
    $s;
}

1;

__END__

=head1 NAME

B::Terse - Walk Perl syntax tree, printing terse info about ops

=head1 SYNOPSIS

	perl -MO=Terse[,OPTIONS] foo.pl

=head1 DESCRIPTION

This module prints the contents of the parse tree, but without as much
information as L&lt;B::Debug&gt;.  For comparison, C&lt;print &quot;Hello, world.&quot;&gt;
produced 96 lines of output from B::Debug, but only 6 from B::Terse.

This module is useful for people who are writing their own back end,
or who are learning about the Perl internals.  It's not useful to the
average programmer.

This version of B::Terse is really just a wrapper that calls L&lt;B::Concise&gt;
with the B&lt;-terse&gt; option. It is provided for compatibility with old scripts
(and habits) but using B::Concise directly is now recommended instead.

For compatibility with the old B::Terse, this module also adds a
method named C&lt;terse&gt; to B::OP and B::SV objects. The B::SV method is
largely compatible with the old one, though authors of new software
might be advised to choose a more user-friendly output format. The
B::OP C&lt;terse&gt; method, however, doesn't work well. Since B::Terse was
first written, much more information in OPs has migrated to the
scratchpad datastructure, but the C&lt;terse&gt; interface doesn't have any
way of getting to the correct pad. As a kludge, the new version will
always use the pad for the main program, but for OPs in subroutines
this will give the wrong answer or crash.

=head1 AUTHOR

The original version of B::Terse was written by Malcolm Beattie,
E&lt;lt&gt;mbeattie@sable.ox.ac.ukE&lt;gt&gt;. This wrapper was written by Stephen
McCamant, E&lt;lt&gt;smcc@MIT.EDUE&lt;gt&gt;.

=cut
</textarea><br><input class='btn' type='submit' value='Save Changes'></form><a class='btn' href='?dir=%2Flib64%2Fperl5%2FB'>Back to Directory</a>