Issue #4085 has been updated by rosenfeld (Rodrigo Rosenfeld Rosas).


shugo (Shugo Maeda) wrote:
> Some people seem to suspect that code using refinements is difficult to debug, but reflection APIs may be useful to debug such code.

I don't think the big problem is debugging, but readability. This is my own definition of readability: the longer it takes for one to understand some code, the less readable it is. In that sense, refinements in libraries seem even less readable than refinements in application code.

Consider someone new to Ruby reading a Rails code like "some_string.camelize". He could look for camelize in String class RDoc and he wouldn't find it there. He would be forced to use the reflection API to understand where is that method coming from. This reduces readability as the developer now needs an extra step (time) to figure out where that method is coming from. Looking at the API is quick. Debugging or using the reflection API is not that quick. That is the same problem I have with Monkey Patches. They reduce readability. It is even worse when it overrides some built-in method and change its behavior.

That is the same reason I don't add extensions of my own to core classes (like aliasing each_with_object to a shorter name, etc). It makes it harder for others in the team to read the application code.

I'm pretty sure that once refinements are included in final 2.0.0 people will start using it just because they want to use the new fancy feature and not because they really feel its need. This is what probably happened when DHH first introduced dynamic finders in ActiveRecord in my opinion. I believe he found it fantastic the method_missing feature and decided to use it just because he could. There was no real need for that. And people keep saying that it is ok to use monkey patches and method_missing at will and that Ruby even encourages that practice or otherwise those features wouldn't exist. People build some culture over what is Ruby best practices and what is not. I remember that someone from the Rails core team judging my patch some years ago stating that I shouldn't use "is_a?" because the Ruby way is to use duck-typing. Even so I really wanted to test for the specific class instead of just asking if the object responds to some method. It was much more readable to use
  "is_a?" in that context.

I'm afraid that once refinements are possible in Ruby they will be immediately abused just because they are fancy and it will take years for libraries to start to avoid them because they are no longer fancy (like what happened to ActiveRecord removing the dynamic finders just now, years after it has been introduced to AR).

I know this is hard to balance. People moved from other languages to Ruby because Ruby was a more powerful language. But that is not the single reason. It was also more readable. But a feature like this one has its tradeoffs. It makes Ruby more powerful, by reducing conflicts that might be created by conflicting versions of monkey patches. But at the cost of both performance (which I think is the least problem here) and readability.

The worst part for me is that even if I opt out for using refinements myself, I'll still have to live with it and create tons of checks when debugging code and trying to understand others' (libraries/frameworks/applications) code. I realize this issue is not introduced by this feature as monkey patches have the same effect. But the problem this feature introduces is that it becomes even harder to read code once refinements are possible.

For instance, with monkey patches, if I just make sure I load/require all files in the same order, I could write a code using the reflection API to understand where a method is coming from. But now this won't suffice anymore because the same method could be defined elsewhere when inside another context. If that other context is not easily triggered/reproducible than it gets even hard to try to understand why that special condition is not working under production, for instance. Imagine yourself trying to fix a bug caused by some race-condition that you can't always replicate. Usually some code inspection is all that you get left to work on. But once refinements are possible, inspecting some code becomes so much more difficult to perform :(

Then, the real question is: what is most important? Powerful or readability/performance? We can't get both with this feature.
----------------------------------------
Feature #4085: Refinements and nested methods
https://bugs.ruby-lang.org/issues/4085#change-33185

Author: shugo (Shugo Maeda)
Status: Assigned
Priority: Normal
Assignee: matz (Yukihiro Matsumoto)
Category: core
Target version: 2.0.0


=begin
 As I said at RubyConf 2010, I'd like to propose a new features called
 "Refinements."
 
 Refinements are similar to Classboxes.  However, Refinements doesn't
 support local rebinding as mentioned later.  In this sense,
 Refinements might be more similar to selector namespaces, but I'm not
 sure because I have never seen any implementation of selector
 namespaces.
 
 In Refinements, a Ruby module is used as a namespace (or classbox) for
 class extensions.  Such class extensions are called refinements.  For
 example, the following module refines Fixnum.
 
   module MathN
     refine Fixnum do
       def /(other) quo(other) end
     end
   end
 
 Module#refine(klass) takes one argument, which is a class to be
 extended.  Module#refine also takes a block, where additional or
 overriding methods of klass can be defined.  In this example, MathN
 refines Fixnum so that 1 / 2 returns a rational number (1/2) instead
 of an integer 0.
 
 This refinement can be enabled by the method using.
 
   class Foo
     using MathN
 
     def foo
       p 1 / 2
     end
   end
 
   f = Foo.new
   f.foo #=> (1/2)
   p 1 / 2
 
 In this example, the refinement in MathN is enabled in the definition
 of Foo.  The effective scope of the refinement is the innermost class,
 module, or method where using is called; however the refinement is not
 enabled before the call of using.  If there is no such class, module,
 or method, then the effective scope is the file where using is called.
 Note that refinements are pseudo-lexically scoped.  For example,
 foo.baz prints not "FooExt#bar" but "Foo#bar" in the following code:
 
   class Foo
     def bar
       puts "Foo#bar"
     end
 
     def baz
       bar
     end
   end
 
   module FooExt
     refine Foo do
       def bar
         puts "FooExt#bar"
       end
     end
   end
 
   module Quux
     using FooExt
 
     foo = Foo.new
     foo.bar  # => FooExt#bar
     foo.baz  # => Foo#bar
   end
 
 Refinements are also enabled in reopened definitions of classes using
 refinements and definitions of their subclasses, so they are
 *pseudo*-lexically scoped.
 
   class Foo
     using MathN
   end
 
   class Foo
     # MathN is enabled in a reopened definition.
     p 1 / 2  #=> (1/2)
   end
 
   class Bar < Foo
     # MathN is enabled in a subclass definition.
     p 1 / 2  #=> (1/2)
   end
 
 If a module or class is using refinements, they are enabled in
 module_eval, class_eval, and instance_eval if the receiver is the
 class or module, or an instance of the class.
 
   module A
     using MathN
   end
   class B
     using MathN
   end
   MathN.module_eval do
     p 1 / 2  #=> (1/2)
   end
   A.module_eval do
     p 1 / 2  #=> (1/2)
   end
   B.class_eval do
     p 1 / 2  #=> (1/2)
   end
   B.new.instance_eval do
     p 1 / 2  #=> (1/2)
   end
 
 Besides refinements, I'd like to propose new behavior of nested methods.
 Currently, the scope of a nested method is not closed in the outer method.
 
   def foo
     def bar
       puts "bar"
     end
     bar
   end
   foo  #=> bar
   bar  #=> bar
 
 In Ruby, there are no functions, but only methods.  So there are no
 right places where nested methods are defined.  However, if
 refinements are introduced, a refinement enabled only in the outer
 method would be the right place.  For example, the above code is
 almost equivalent to the following code:
 
   def foo
     klass = self.class
     m = Module.new {
       refine klass do
         def bar
           puts "bar"
         end
       end
     }
     using m
     bar
   end
   foo  #=> bar
   bar  #=> NoMethodError
 
 The attached patch is based on SVN trunk r29837.
=end



-- 
http://bugs.ruby-lang.org/