Using marc4j from jruby

marc4j_jruby.rb

# I just nabbed the source of marc4j and built it with "ant jar"

require 'marc4j-2.5.1-beta.jar'

# Conveniently add Enumerable to the reader interface so I can get #each, #each_with_index, etc.
# This would be automatic if MarcReader were specified as an iterable, as per a recent github issue
# on the marc4j repo (https://github.com/marc4j/marc4j/issues/11)

module org.marc4j::MarcReader
  include Enumerable
  def each
    if block_given?
      while self.hasNext
        yield self.next
      end
    else
      self.to_enum(:each)
    end
  end  
end


# Pull in the 'batch.dat' file from the ruby-marc test suite
istream = java.io.FileInputStream.new('batch.dat')

# Make a reader out of it. I'm specifying utf-8, but you could leave it blank
# and get the "best guess" as well
reader = org.marc4j.MarcStreamReader.new(istream, 'UTF-8')

reader.each do |r|
  # whatever
end


# Do the same thing with a permissive reader
istream = java.io.FileInputStream.new('batch.dat')
reader = org.marc4j.MarcPermissiveStreamReader.new(istream, true, true)
iter = reader.each # get the iterator
puts iter.next