Here is the code. Its a 254 line function :).

TAG - this is an object that contains the rules for controlling whether we can indent or not. It also provides one or two other convenience methods.

$mod - if it isn't obvious, this controls our indentation level so the more child tags we have the more we indent (using tabs). 
sub prepareOutput
{
    my $self = shift;
    my $node = shift;
    
    unless(defined($node))
    {
        warn "No parsable node or tree passed.";
        return;
    }
    
    # Add the current node tag to the stack
    $current_tag = $node->tag;
    unshift(@{$old_tag},$current_tag);
    
    # See if we are a frameset
    if($node->tag eq 'frameset')
    {
        $self->[FRAMESET]++;
    }
    
    if(DEBUG)
    {
        $self->[OUTPUT] .= $old_parent->[0] if defined($old_parent->[0
+]);
        $self->[OUTPUT] .= $old_tag->[0] if defined($old_tag->[0]);
    }
    
    # If we are a child
    if(defined($node->parent))
    {
        # Store the parent in the parent tag - we do this incase 
        # we meet some text and we don't have the parent call availabl
+e
        $current_parent = $node->parent->tag;
        unshift(@{$old_parent},$current_parent);

        # Check that the parent allows us to be indented, accomodating
        # for comments.
        if(
            $current_parent eq 'td' 
        &&
            $node->parent->content_list < 1
        &&
            $node->tag eq 'a' || $node->tag eq 'img'
        )
        {
            $self->[OUTPUT] .= '<' . $node->tag;
        }
        elsif($self->[TAG]->canIndentChild($current_parent))
        {
            if($node->tag eq '~comment')
            {
                $self->[OUTPUT] .= "\t" x $mod;
                $self->[OUTPUT] .= '<!--';
            }
            else
            {
                $self->[OUTPUT] .= "\t" x $mod;
                $self->[OUTPUT] .= '<' . $node->tag;
            }
        }
        else
        {
            if($node->tag eq '~comment')
            {
                $self->[OUTPUT] .= '<!--';
            }
            else
            {
                $self->[OUTPUT] .= '<' . $node->tag;
            }            
        }
    }
    else
    {
        # No parent, print
        if($node->tag eq '~comment')
        {
            $self->[OUTPUT] .= "\t" x $mod;
            $self->[OUTPUT] .= '<!--';
        }
        else
        {
            $self->[OUTPUT] .= "\t" x $mod;
            $self->[OUTPUT] .= '<' . $node->tag;
        }
    }
    
    # Print the attributes for this tag, also takes care of space pres
+erving
    # for XHTML
    $self->printAttributes($node);
        
    # Determine how we end this tag, ie if it is a terminal like <br> 
+or <hr>
    my $is_terminal = 0;
    if($self->[TAG]->isTerminal($node->tag))
    {
        if($node->tag eq '~comment')
        {
            $self->[OUTPUT] .= ' --';
        }
        else
        {
            $self->[OUTPUT] .= ' /' if $self->[CONFIG]->setting('xhtml
+_output') eq 'yes';
        }
        $is_terminal = 1;
    }
    
    # This accomodates whether we indent future nodes or keep the 'tig
+ht'
    my @list = $node->content_list();
    #print $list[0],"\n";
    
    if(
        defined($current_parent) && $current_parent eq 'td' 
        &&
            $node->parent->content_list < 1
        &&
            $node->tag eq 'a' || $node->tag eq 'img'
    )
    {
        $self->[OUTPUT] .= ">";
    }
    elsif(
            $node->tag eq 'td'
        &&
            $node->content_list == 1
        &&
            defined($list[0]) && ref($list[0])
        &&
            $list[0]->tag eq 'img'
    )
    {
        $self->[OUTPUT] .= '>';
    }
    elsif(
            $node->tag eq 'td'
        &&
            $node->content_list > 1
        &&
            defined($list[0]) && ref($list[0])
        &&
            $list[0]->tag eq 'img'
    )
    {
        $self->[OUTPUT] .= ">\n" . ("\t" x ($mod + 1));
    }
    elsif($self->[TAG]->canIndentChild($old_parent->[0]) && ($self->[T
+AG]->canIndentChild($old_tag->[0])))
    {
        $self->[OUTPUT] .= ">\n";
    }
    else
    {
        $self->[OUTPUT] .= ">";
    }

    # Now traverse child nodes via recursion or, if the next bit is te
+xt,
    # print that out.
    $mod++;
    my @nodes = $node->content_list;
    my $n_count = @nodes;

#    foreach my $c ($node->content_list) 
    my $c;
    for(my $i = 0; $i < $n_count; $i++)
    {
        $c = $nodes[$i];
        next unless defined($c);
        if(ref $c)
        {
            if($i > 0 && (ref($nodes[$i - 1]) && $nodes[$i - 1]->tag e
+q 'a'))
            {
                $self->[OUTPUT] .= "\n";
            }
            $self->prepareOutput($c);
        }
        elsif($i > 0 && (ref($nodes[$i - 1]) && $nodes[$i - 1]->tag eq
+ 'a'))
        {
            # Its an anchor followed by some punctuation. We need the 
+pnuctuation
            # on the same line for IE.
            
            if($c =~ /^[.,);?\:]/)
            {
                $self->[OUTPUT] .= substr($c, 0, 1);
                $c = substr($c, 1, length($c));
            }
            $self->[OUTPUT] .= "\n";
            $self->printNodeText($c);
        }
        else
        {
            $self->printNodeText($c);
        }
    }
    $mod--;
    
    # Just in case we had an anchor at the end and need to indent
    if(ref $c && $c->tag eq 'a' && $node->tag =~ /p|td/)
    {
        $self->[OUTPUT] .= "\n";
    }
    
    # If our tag was not a terminal, we must $self->[OUTPUT] .= the en
+d tag, taking care
    # to ensure we have followed our indentation rules.
    if(!$is_terminal)
    {
        if(DEBUG)
        {
            $self->[OUTPUT] .= $old_parent->[0] if defined($old_parent
+->[0]);
            $self->[OUTPUT] .= $old_tag->[0] if defined($old_tag->[0])
+;
        }
        
        if($node->tag eq 'a')
        {
            $self->[OUTPUT] .= '</a>';
        }
        elsif(
            $node->tag eq 'td'
            &&
                $node->content_list < 1
        )
        {
            $self->[OUTPUT] .= '</' . $node->tag . ">\n";
        }
        elsif(
            $node->tag eq 'td'
            &&
                $node->content_list == 1
            &&
                defined($list[0]) && ref($list[0])
            &&
                $list[0]->tag eq 'img'
        )
        {
            $self->[OUTPUT] .= '</' . $node->tag . ">\n";
        }
        elsif($self->[TAG]->canIndentChild($old_tag->[0]))
        {
            ($node->tag eq '~comment') ? $self->[OUTPUT] .= "\t" x $mo
+d . " -->\n" :    $self->[OUTPUT] .= "\t" x $mod . '</' . $node->tag 
+. ">\n";
        }
        else
        {
            if($self->[TAG]->canIndentChild($old_parent->[0]))
            {
                ($node->tag eq '~comment') ? $self->[OUTPUT] .= " -->\
+n": $self->[OUTPUT] .= '</' . $node->tag . ">\n";
            }
            else
            {
                ($node->tag eq '~comment') ? $self->[OUTPUT] .= " -->"
+: $self->[OUTPUT] .= '</' . $node->tag . ">";
            }
        }
    }
    
    # Remove our current bits off the stacks and return.
    $current_parent = shift( @{$old_parent});
    $current_tag = shift( @{$old_tag});
}

[download]

In reply to Output stage code (long) by simon.proctor
in thread Output of HTML tree built with TreeBuilder by simon.proctor

Title:
Use:  <p> text here (a paragraph) </p>
and:  <code> code here </code>
to format your post, it's "PerlMonks-approved HTML":


  • Posts are HTML formatted. Put <p> </p> tags around your paragraphs. Put <code> </code> tags around your code and data!
  • Titles consisting of a single word are discouraged, and in most cases are disallowed outright.
  • Read Where should I post X? if you're not absolutely sure you're posting in the right place.
  • Please read these before you post! —
  • Posts may use any of the Perl Monks Approved HTML tags:
    a, abbr, b, big, blockquote, br, caption, center, col, colgroup, dd, del, details, div, dl, dt, em, font, h1, h2, h3, h4, h5, h6, hr, i, ins, li, ol, p, pre, readmore, small, span, spoiler, strike, strong, sub, summary, sup, table, tbody, td, tfoot, th, thead, tr, tt, u, ul, wbr
  • You may need to use entities for some characters, as follows. (Exception: Within code tags, you can put the characters literally.)
            For:     Use:
    & &amp;
    < &lt;
    > &gt;
    [ &#91;
    ] &#93;
  • Link using PerlMonks shortcuts! What shortcuts can I use for linking?
  • See Writeup Formatting Tips and other pages linked from there for more info.