Ruby分片读取文件

有两个办法

第一， 使用linux的split命令，分割文件，然后在逐个读取文件

		#first cut file into 50k byte (51200) slice, then send each slice using resume
		#
		# we accept file max size 500M now. should we support larger file, change -a
		def split
			prefix = "chunk#{@session_id}_"
			cmd  = "split -b #{@slice_block} -d -a 4 #{@file_path}  #{prefix}"
			puts "cmd is #{cmd}"
			`#{cmd}` #shell to split files
			raise Exception,  "fail to split file, command as split -b #{@slice_block} -d -a 4 #{@file_file} #{prefix}" if $?.to_i != 0
		end

读取的时候，可以使用如下方法

file_chunks = `ls chunk#{@session_id}_*`.split("\n")
			file_chunks.each { |chunk|

... ...
}

第二， 扩展Ruby的File， 增加each_chunk方法

class File
  MEGABYTE = 300 * 1024

  def each_chunk(chunk_size=MEGABYTE)
    yield read(chunk_size) until eof?
  end
end