Forum: Ruby serializing Class and Module objects

Announcement (2017-05-07): www.ruby-forum.com is now read-only since I unfortunately do not have the time to support and maintain the forum any more. Please see rubyonrails.org/community and ruby-lang.org/en/community for other Rails- und Ruby-related community platforms.
Joel VanderWerf (Guest)
on 2006-02-03 22:12
(Received via mailing list)
Hey, _why, what do you think about allowing YAML to dump/load Class and
Module objects rather than gagging? I've posted code like the following
(different versions for 1.8.2 and 1.8.4) several times in response to
inquiries on ruby-talk, and found it useful myself.



=====================================================
Test code:

  enum_y = [Enumerable, Comparable, String, File].to_yaml
  puts enum_y
  p YAML.load(enum_y)

=====================================================
Test output:

---
- !ruby/module Enumerable
- !ruby/module Comparable
- !ruby/class String
- !ruby/class File
[Enumerable, Comparable, String, File]

=====================================================
require 'yaml'

if defined?(YAML.type_tag) # old version of YAML

  class Module
    def is_complex_yaml?
      false
    end
    def to_yaml( opts = {} )
      YAML::quick_emit( nil, opts ) { |out|
        out << "!ruby/module "
        self.name.to_yaml( :Emitter => out )
      }
    end
  end
  YAML.add_ruby_type(/^module/) do |type, val|
    subtype, subclass = YAML.read_type_class(type, Module)
    val.split(/::/).inject(Object) { |p, n| p.const_get(n)}
  end

  class Class
    def to_yaml( opts = {} )
      YAML::quick_emit( nil, opts ) { |out|
        out << "!ruby/class "
        self.name.to_yaml( :Emitter => out )
      }
    end
  end
  YAML.add_ruby_type(/^class/) do |type, val|
    subtype, subclass = YAML.read_type_class(type, Class)
    val.split(/::/).inject(Object) { |p, n| p.const_get(n)}
  end

else

  class Module
    yaml_as "tag:ruby.yaml.org,2002:module"

    def Module.yaml_new( klass, tag, val )
      if String === val
        val.split(/::/).inject(Object) {|m, n| m.const_get(n)}
      else
        raise YAML::TypeError, "Invalid Module: " + val.inspect
      end
    end

    def to_yaml( opts = {} )
      YAML::quick_emit( nil, opts ) { |out|
        out.scalar( "tag:ruby.yaml.org,2002:module", self.name, :plain )
      }
    end
  end

  class Class
    yaml_as "tag:ruby.yaml.org,2002:class"

    def Class.yaml_new( klass, tag, val )
      if String === val
        val.split(/::/).inject(Object) {|m, n| m.const_get(n)}
      else
        raise YAML::TypeError, "Invalid Class: " + val.inspect
      end
    end

    def to_yaml( opts = {} )
      YAML::quick_emit( nil, opts ) { |out|
        out.scalar( "tag:ruby.yaml.org,2002:class", self.name, :plain )
      }
    end
  end

end
why the lucky stiff (Guest)
on 2006-02-03 22:25
(Received via mailing list)
Joel VanderWerf wrote:

>Hey, _why, what do you think about allowing YAML to dump/load Class and
>Module objects rather than gagging? I've posted code like the following
>(different versions for 1.8.2 and 1.8.4) several times in response to
>inquiries on ruby-talk, and found it useful myself.
>
>
I've been reluctant, because reloading these objects causes trouble if
you haven't required the right libraries.  But I think if YAML's error
message was okay, it would work.

Oh and the best place for getting Syck requests through is now here:
http://code.whytheluckystiff.net/syck/newticket

_why
Joel VanderWerf (Guest)
on 2006-02-03 22:34
(Received via mailing list)
why the lucky stiff wrote:
> message was okay, it would work.
I thought about rescuing and re-raising the exception as a different
exception class, but maybe it is better to leave it as it is:

NameError: uninitialized constant Foo

This is the same exception you would get if you loaded a script file.

The way Marshal does it (in the case of a dumped class that is loaded
without having the class definition required) is:

ArgumentError: undefined class/module Foo

Maybe that's how YAML should do it for consistency?

I'd be happy with either way.

> Oh and the best place for getting Syck requests through is now here:
> http://code.whytheluckystiff.net/syck/newticket

Thanks!
Joel VanderWerf (Guest)
on 2006-02-11 21:52
(Received via mailing list)
why the lucky stiff wrote:
> message was okay, it would work.
>
> Oh and the best place for getting Syck requests through is now here:
> http://code.whytheluckystiff.net/syck/newticket

Ok. I've submitted this patch and also the :SortKeys issue to the Syck
Trak.
This topic is locked and can not be replied to.