Forum: IronRuby Code Review: YamlRefactoring1

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.
Cb51033949ffccd982ae32c9f890f25a?d=identicon&s=25 Tomas Matousek (Guest)
on 2008-12-03 21:17
(Received via mailing list)
Attachment: YamlRefactoring1.diff (100 KB)
tfpt review "/shelveset:YamlRefactoring1;REDMOND\tomat"
  Comment  :
  Various Yaml related changes:
  -     Integer and Numeric now derives from RubyObject so that a new
CLR type is not created when these classes are subclassed.
  -     All classes derived from RubyObject need to implement a
deserializing constructor.
  -     If a class in C# library represented a Ruby class and also
implemented its methods using RubyMethod attribute the static
[RubyMethod] methods were visible on the singleton class of that class.
This caused problems with method lookup. This shelveset adds a flag to
the class loader that hides all CLR methods from a class that implements
a Ruby class (via RubyClass attribute). This fix removes the necessity
of splitting of classes into "Ops" and "real classes" artificially (an
example of which is YAML::Stream).
  -     Refactors RubyTypeBuilder.DefineConstructors a little bit and
fixes problem with constructors taking RubyClass - the generated code
was unverifiable (an instruction was missing).
  -     Removes mscorlib dependency from Yaml tests so that they can be
run by MRI as well.

Tomas
C98ff02b6d5c06f185a30ed1a69db766?d=identicon&s=25 Oleg Tkachenko (Guest)
on 2008-12-03 22:42
(Received via mailing list)
Looks great!

--
Oleg
This topic is locked and can not be replied to.