Question

This is what I want to do:

Remove “span” nodes with a class of “none”.

Remove “extra” nodes but keep the text inside them.

Remove any “br” nodes and replace them with “p” nodes

<p class="normal">
    <span class="none">
        <extra>Some text goes here</extra>
    </span>
    <span class="none">
        <br/>
    </span>
    <span class="none">
        <extra>Some other text goes here</extra>
        <br/>
    </span>
</p>

This is the output I’d like to achieve:

<p class="normal">Some text goes here</p>
<p class="normal">Some other text goes here</p>

I’ve tried this so far:

doc.xpath('html/body/p/span').each do |span|
    span.attribute_nodes.each do |a|
       if a.value == "none"
          span.children.each do |child|
             span.parent << child
          end
          span.remove
       end
    end
end

But this is the output I’m getting, it’s not even in the right order:

<p class="normal"><br /><br />Some text goes hereSome other text goes here</p>

Answer 1

Try this out

require 'rubygems'
require 'nokogiri'

doc = Nokogiri::XML(DATA)
doc.css("span.none, extra").each do |span|
  span.swap(span.children)
end

# via http://stackoverflow.com/questions/8937846/how-do-i-wrap-html-untagged-text-with-p-tag-using-nokogiri
doc.search("//br/preceding-sibling::text()|//br/following-sibling::text()").each do |node|
  if node.content !~ /\A\s*\Z/
    node.replace(doc.create_element('p', node))
  end
end

doc.css('br').remove

puts doc

__END__
<p class="normal">
    <span class="none">
        <extra>Some text goes here</extra>
    </span>
    <span class="none">
        <br/>
    </span>
    <span class="none">
        <extra>Some other text goes here</extra>
        <br/>
    </span>
</p>

Which prints

<?xml version="1.0"?>
<p class="normal">

        <p>Some text goes here</p>





        <p>Some other text goes here</p>


</p>