pdf_mess.rb

@potential_errors =[]
params = { filepicker: true, created_at: DateTime.parse("2015-01-25 14:48:09")..DateTime.parse("2015-04-01 19:50") }
grouped_count = Attachment.where(params).group("original_filename").having("count(1) > 1").count ; nil
grouped_attachments = grouped_count.map { |count| Attachment.where(params.merge(original_filename: count.first)).to_a } ; nil

grouped_attachments.each do |group|
  arr = group.map &:document_file_size
  if arr.uniq != arr
    @potential_errors << group
  end
end ; nil
puts "#{@potential_errors.count} groups"
puts "#{@potential_errors.flatten.count} attachments"

@potential_errors.flatten! ; nil
@potential_errors.reject! { |att| att.subject.bill_status.read_attribute(:status) == "done" } ; nil

puts "#{@potential_errors.count} attachments remaining"
@att_errors = []
@potential_errors.each_with_index do |att, i|
  puts "counter: #{i + 1}"
  begin
    # perceived issue with update_attributes! (could be my imagination)
    att.document = att.original_document
    att.save!
  rescue => e
    @att_errors << [e.message, att.id, att.subject_id]
  end
end ; nil

#
# The code from here down hasn't been used for a couple of days but may nevertheless prove relevant
#
@csv_errors = []
@potential_csv_errors = []
@potential_errors.each_with_index do |group, i|
  puts "Working on group #{i +1}"
  begin
    group.each do |att|
      bill = att.subject
      unless bill.bill_status.status == "done"
        bill.submissions.each do |bs|
          @potential_csv_errors << [att.id, bill.id, att.report_type.description, att.original_filename, att.document_file_size, att.document_file_name, bill.bill_status.status, bill.date_of_service, bs.id, bs.created_at, bs.updated_at]
        end
      end
    end
  rescue => e
    @csv_errors << [e, group]
  end
end ; nil

potential_errors_file = CSV.generate do |csv|
  csv << ["Attachment ID", "Bill ID", "Report Type", "Original Filename", "Document File Size", "Document Filename", "Bill Status", "Date of Service", "Submission ID", "Submission Created", "Submission Updated"]
  @potential_csv_errors.each { |x| csv << x }
end

@errors = []
@with_errors = []
grouped_attachments.each_with_index do |group, i|
  puts i
  group.each_with_index do |attachment, index|
    begin
      bill = attachment.subject
      group[(index)..-1].each do |remaining_attachment|
        bill.submissions.map do |bs|
          if bs.created_at > remaining_attachment.created_at && bill.bill_status.status != "done"
            @with_errors << [attachment.id, attachment.subject_id, bs.id, bs.created_at]
          end
        end
      end
     rescue => e
       @errors << [e, group]
    end
  end
end ; nil

column_names = ["Attachment ID", "Bill ID", "Bill Submission ID", "Bill Submission Created At"]
CSV.generate do |csv|
  csv << column_names
  @with_errors.each do |err|
    csv << err
  end
end