gistfile1.txt

require 'sanitize'

STRIPPERS = %w(br p).freeze

def cleaner
  lambda do |env|
    return nil if !STRIPPERS.include?(env[:node_name])
    n = env[:node]
    txt = Nokogiri::XML::Text.new(' ', n.document)
    n.children.each do |c|
      txt.add_child(c)
    end
    txt.add_child(Nokogiri::XML::Text.new(' ', n.document))
    { :node => txt }
  end
end

f = '<img src="http://jobs.43folders.com/files/company_logos/823290/dd98f0e1.png" align="right" />AT&#38;T/Milpitas, CA<br/><br/>Sr Technical Architect<br />'
f.each_line do |line|
  result = Sanitize.clean(line, :remove_contents => %w(script style), :transformers => cleaner)
  # Space character class is necessary as \s is not Unicode-aware in Ruby 1.9.1 anymore.
  puts result.gsub(/\p{Space}+/u, ' ')  
end