源码阅读笔记

autoload

autoload可以加载（注册）任何模块的类或常量，不能有::操作符的...
特别广泛应用在如下代码结构：

#rack.rb
#这样的文件结构ruby源码里随处可见
module Rack
  autoload :Builder, "rack/builder"
  autoload :Cascade, "rack/cascade"
  autoload :Chunked, "rack/chunked"
  autoload :CommonLogger, "rack/commonlogger"
  autoload :ConditionalGet, "rack/conditionalget"
  autoload :Config, "rack/config"
  ...
end
autoload本质是会调用Kernel.require,但是又和require有区别。
可以说autoload是一个smart的require...比require更加智能灵活。Just-in-Time..
几个简单的例子：

要被require的文件，mylibrary.rb
puts "I was loaded!"

class MyLibrary
end

require mylibrary时文件立即执行。

irb(main):001:0> require 'mylibrary'
I was loaded!
=> true




使用autoload，只有使用到需要的常量或类文件才被加载。。我们真正需要用某个文件时才加载，而require是直接加载，不管你是否会用到。

irb(main):001:0> autoload :MyLibrary, 'mylibrary'
=> nil
irb(main):002:0> MyLibrary.new
I was loaded!
=> #<MyLibrary:0x0b1jef>



其实，我觉得require更实用一些。。预先一次性加载来的痛快嘛。。

这个作者举了一个实例的例子来说明autoload的应用场景：http://www.subelsky.com/2008/05/using-rubys-autoload-method-to.html

link: http://www.rubyinside.com/ruby-techniques-revealed-autoload-1652.html
link: http://www.ruby-doc.org/core/classes/Kernel.html#M005968
link: http://www.ruby-forum.com/topic/172385
link: http://www.subelsky.com/2008/05/using-rubys-autoload-method-to.html

 


trap

trap("SIGINT") do
  irb.signal_handle
end

捕捉信号量。然后执行block

Signal.trap(0, proc { puts "Terminating: #{$$}" })
Signal.trap("CLD")  { puts "Child died" }

test:

#     test(int_cmd, file1 [, file2] ) => obj
  #
  #
  # Uses the integer <i>aCmd</i> to perform various tests on
  # <i>file1</i> (first table below) or on <i>file1</i> and
  # <i>file2</i> (second table).
  #
  # File tests on a single file:
  #
  #   Test   Returns   Meaning
  #    ?A  | Time    | Last access time for file1
  #    ?b  | boolean | True if file1 is a block device
  #    ?c  | boolean | True if file1 is a character device
  #    ?C  | Time    | Last change time for file1
  #    ?d  | boolean | True if file1 exists and is a directory
  #    ?e  | boolean | True if file1 exists
  #    ?f  | boolean | True if file1 exists and is a regular file
  #    ?g  | boolean | True if file1 has the \CF{setgid} bit
  #        |         | set (false under NT)
  #    ?G  | boolean | True if file1 exists and has a group
  #        |         | ownership equal to the caller's group
  #    ?k  | boolean | True if file1 exists and has the sticky bit set
  #    ?l  | boolean | True if file1 exists and is a symbolic link
  #    ?M  | Time    | Last modification time for file1
  #    ?o  | boolean | True if file1 exists and is owned by
  #        |         | the caller's effective uid
  #    ?O  | boolean | True if file1 exists and is owned by
  #        |         | the caller's real uid
  #    ?p  | boolean | True if file1 exists and is a fifo
  #    ?r  | boolean | True if file1 is readable by the effective
  #        |         | uid/gid of the caller
  #    ?R  | boolean | True if file is readable by the real
  #        |         | uid/gid of the caller
  #    ?s  | int/nil | If file1 has nonzero size, return the size,
  #        |         | otherwise return nil
  #    ?S  | boolean | True if file1 exists and is a socket
  #    ?u  | boolean | True if file1 has the setuid bit set
  #    ?w  | boolean | True if file1 exists and is writable by
  #        |         | the effective uid/gid
  #    ?W  | boolean | True if file1 exists and is writable by
  #        |         | the real uid/gid
  #    ?x  | boolean | True if file1 exists and is executable by
  #        |         | the effective uid/gid
  #    ?X  | boolean | True if file1 exists and is executable by
  #        |         | the real uid/gid
  #    ?z  | boolean | True if file1 exists and has a zero length
  #
  # Tests that take two files:
  #
  #    ?-  | boolean | True if file1 and file2 are identical
  #    ?=  | boolean | True if the modification times of file1
  #        |         | and file2 are equal
  #    ?<  | boolean | True if the modification time of file1
  #        |         | is prior to that of file2
  #    ?>  | boolean | True if the modification time of file1
  #        |         | is after that of file2
  #
  #
  def test(int_cmd, file1, file2 )
    # This is just a stub for a builtin Ruby method.
    # See the top of this file for more info.
  end

irb(main):045:0> test("A", "/var/log/mysql.log")
=> Fri Apr 02 10:22:46 +0800 2010
irb(main):046:0> test("C", "/var/log/mysql.log")
=> Fri Apr 02 10:22:46 +0800 2010
irb(main):047:0> test("c", "/var/log/mysql.log")
=> false
irb(main):048:0> test("e", "/var/log/mysql.log")
=> true

这个可以用来调试。。很强大：

set_trace_func proc { |event, file, line, id, binding, classname|
  #        printf "%8s %s:%-2d %10s %8s\n", event, file, line, id, classname
  #     }

rackup http://www.elanso.com/ArticleModule/sourcearticle.aspx?idx=L9IYPUNOVwLmP0KUG9SEKAIi