gistfile1.rb

# Given a set of documents this method returns a list of tags associated with
# those documents ordered by the ones occuring on the most documents. Tags th
# only appear on one doc are excluded from the list.
# If the user supplies a specific tag to exclude it will not be included
# in the list.
def self.related_tags(docs, exclude = nil)
  doc_ids = docs[0, 20].map { |doc| doc.id }.join(',')
  tags = self.connection.select_all(
    "SELECT word_tag_id, COUNT(word_tag_id) FROM word_documents_word_tags
      WHERE word_document_id IN (#{doc_ids}) GROUP BY word_tag_id
      HAVING COUNT(*) > 1 ORDER BY COUNT(*) DESC LIMIT 26")
  tag_ids = tags.map { |tag| tag['word_tag_id'] }
  related = WordTag.find(:all, :conditions => { :id => tag_ids })
  related.reject! { |tag| tag.name == exclude } if exclude
  related
end