Question

I have a regular xml object created from a response of a web service.

I need to get some specific values from some specific keys… for example:

<tag>
 <tag2>
  <tag3>
   <needThisValue>3</needThisValue>
   <tag4>
    <needThisValue2>some text</needThisValue2>
   </tag4>
  </tag3>
 </tag2>
</tag>

How can I get <needThisValue> and <needThisValue2> in Ruby?

Answer 1

I’m a big fan of Nokogiri:

xml = <<EOT
<tag>
 <tag2>
  <tag3>
   <needThisValue>3</needThisValue>
   <tag4>
    <needThisValue2>some text</needThisValue2>
   </tag4>
  </tag3>
 </tag2>
</tag>
EOT

This creates a document for parsing:

require 'nokogiri'
doc = Nokogiri::XML(xml)

Use at to find the first node matching the accessor:

doc.at('needThisValue2').class # => Nokogiri::XML::Element

Or search to find all nodes matching the accessor as a NodeSet, which acts like an Array:

doc.search('needThisValue2').class # => Nokogiri::XML::NodeSet
doc.search('needThisValue2')[0].class # => Nokogiri::XML::Element

This uses a CSS accessor to locate the first instance of each node:

doc.at('needThisValue').text # => "3"
doc.at('needThisValue2').text # => "some text"

Again with the NodeSet using CSS:

doc.search('needThisValue')[0].text # => "3"
doc.search('needThisValue2')[0].text # => "some text"

You can use XPath accessors instead of CSS if you want:

doc.at('//needThisValue').text # => "3"
doc.search('//needThisValue2').first.text # => "some text"

Go through the tutorials to get a jumpstart. It’s very powerful and quite easy to use.