UMI ETD XML to MARCXML with Ruby

gistfile1.rb

#!/usr/bin/env ruby

require 'nokogiri'

# where the files at
my_files = Dir.chdir(ARGV[0]) { Dir.glob("./*").map{|x| File.expand_path(x) } }

builder = Nokogiri::XML::Builder.new(:encoding => 'UTF-8') do |xml|
	xml.collection(:xmlns => 'http://www.loc.gov/MARC21/slim') {
		my_files.each do |my_file|
			my_file = Nokogiri::XML(File.open(my_file))
			author = my_file.at_xpath("//DISS_author[@type='primary']/DISS_name")
			title = my_file.at_xpath("//DISS_description/DISS_title")
			xml.record {
				xml.datafield(:tag => "100", :ind1 => "1", :ind2 => " ") {
					xml.subfield(:code => "a") { 						
						xml.text "#{author.at_xpath('./DISS_surname').text}, #{author.at_xpath('./DISS_fname').text}"
						xml.text " #{author.at_xpath('./DISS_middle').text}" if author.at_xpath('./DISS_middle')
						xml.text author.at_xpath('./DISS_suffix').inner_text.empty? ? "." : ","
					}
					unless author.at_xpath('./DISS_suffix').inner_text.empty?
						xml.subfield(:code => "c") { 
							xml.text author.at_xpath('./DISS_suffix').text
						}
					end
				}

				xml.datafield(:tag => "245", :ind1 => "1", :ind2 => "0") {
					xml.subfield(:code => "a") { 
						xml.text title.text.split(':',2).first.strip
					}
					if title.text.include?(':')
						xml.subfield(:code => "b") { 
							xml.text title.text.split(':',2).last.strip
						}
					end
					xml.subfield(:code => "c") { 
						xml.text "by #{author.at_xpath('./DISS_fname').text}"
						xml.text " #{author.at_xpath('./DISS_middle').text}"
						xml.text " #{author.at_xpath('./DISS_surname').text}"
						unless author.at_xpath('./DISS_suffix').inner_text.empty?
							xml.text ", #{author.at_xpath('./DISS_suffix').text}"
						end
					}
					xml.subfield(:code => "h") { 
						xml.text "[electronic resource]"
					}
				}
				
				if title.text.include?(':')
					xml.datafield(:tag => "246", :ind1 => "3", :ind2 => "0") {
						xml.subfield(:code => "a") { 
							title_temp = title.text.split(':',2).last.strip
							title_temp[0] = title_temp[0].capitalize
							xml.text title_temp
						}				
					}
				end
				xml.datafield(:tag => "260") {
					xml.subfield(:code => "a") { 
						xml.text "[Fort Worth, Tex.]"
					}
					xml.subfield(:code => "b") { 
						xml.text my_file.at_xpath('//DISS_institution/DISS_inst_name').text
					}
					xml.subfield(:code => "c") { 
						xml.text my_file.at_xpath('//DISS_dates/DISS_comp_date').text
					}			
				}

				xml.datafield(:tag => "490", :ind1 => "0") {
					xml.subfield(:code => "a") { 
						xml.text my_file.at_xpath('//DISS_dates/DISS_comp_date').text
						xml.text " dissertation"
					}	
				}
			
				xml.datafield(:tag => "500") {
					xml.subfield(:code => "a") { 
						xml.text "Title from dissertation title page"
					}	
				}

				unless my_file.at_xpath('//DISS_abstract/DISS_para').text.empty?
					xml.datafield(:tag => "500") {
						xml.subfield(:code => "a") { 
							xml.text "Includes abstract"
						}	
					}
				end

				xml.datafield(:tag => "500") {
					xml.subfield(:code => "a") { 
						xml.text my_file.at_xpath('//DISS_institution/DISS_inst_contact').text						
						xml.text "; advisor, #{my_file.at_xpath('//DISS_advisor/DISS_name/DISS_fname').text}"
						xml.text " #{my_file.at_xpath('//DISS_advisor/DISS_name/DISS_surname').text}"
					}	
				}
				
				my_file.xpath('//DISS_abstract/DISS_para').each do |para_txt|
					xml.datafield(:tag => "520") {
						xml.subfield(:code => "a") { 
							xml.text para_txt.text
						}	
					}
				end

				xml.datafield(:tag => "538") {
					xml.subfield(:code => "a") { 
						xml.text "Mode of access: World Wide Web"
					}	
				}

				if my_file.at_xpath('//DISS_binary/@type').text == 'PDF'
					xml.datafield(:tag => "538") {
						xml.subfield(:code => "a") { 
							xml.text "System requirements: Adobe Acrobat reader"
						}	
					}
				end
			}
		end
	}
end

puts builder.to_xml