Read a sequence file

read-sequence-file-sans-humboldt.rb

def read_sequence_file(path, key_type, value_type)
  uri = "file://#{File.expand_path(path)}"
  conf = Hadoop::Conf::Configuration.new
  fs = Hadoop::Fs::FileSystem.get(java.net.URI.create(uri), conf)
  path = Hadoop::Fs::Path.new(uri)
  reader = Hadoop::Io::SequenceFile::Reader.new(fs, path, conf)
  key = Hadoop::Util::ReflectionUtils.new_instance(reader.key_class, conf)
  value = Hadoop::Util::ReflectionUtils.new_instance(reader.value_class, conf)
  while reader.next(key, value)
    yield key, value
  end
end

read-sequence-file.rb

require 'humboldt'

def read_sequence_file(path)
  uri = "file://#{File.expand_path(path)}"
  conf = Hadoop::Conf::Configuration.new
  fs = Hadoop::Fs::FileSystem.get(java.net.URI.create(uri), conf)
  path = Hadoop::Fs::Path.new(uri)
  reader = Hadoop::Io::SequenceFile::Reader.new(fs, path, conf)
  key = Humboldt::TypeConverter.from_hadoop(reader.key_class.ruby_class).new
  value = Humboldt::TypeConverter.from_hadoop(reader.value_class.ruby_class).new
  while reader.next(key.hadoop, value.hadoop)
    yield key.ruby, value.ruby
  end
end