Bug #21317: Namespaces leak with object IDs

• Status: Closed
• Assignee: -
• Target version: -
• Backport: 3.2: UNKNOWN, 3.3: UNKNOWN, 3.4: UNKNOWN

[fxn (Xavier Noria)]

Namespaces are not transparent for this program

if mod1.object_id == mod2.object_id
  mod1.constants == mod2.constants
end

because it may fail for mod1 = Object; mod2 = ns::Object, which have the same object ID.

[mame (Yusuke Endoh)]

Did you actually test the code? mod1.constants == mod2.constants returns true.

$ cat main.rb
ns = Namespace.new
mod1 = Object
mod2 = ns::Object

p mod1 #=> Object
p mod2 #=> Object
p mod1.object_id == mod2.object_id #=> true
p mod1.constants == mod2.constants #=> true

$ RUBY_NAMESPACE=1 ./miniruby main.rb
./miniruby: warning: Namespace is experimental, and the behavior may change in the future!
See doc/namespace.md for know issues, etc.
Object
Object
true
true

It's great that you're splitting the ticket and discussing each topic in detail. Thank you for that.
That said, when creating tickets, please make sure to include the test code, what you expected to happen, and what actually happened, in the usual format.

[fxn (Xavier Noria)]

I suspect this one was an error of mine. I think we can close.

[fxn (Xavier Noria)]

@byroot (Jean Boussier), if you will, I believe this one can be closed too.

[byroot (Jean Boussier)]

• Status changed from Open to Closed

[Eregon (Benoit Daloze)]

Full reproducer, showing that even though Object is the same reference in all namespaces it can have different constants in different namespaces, but only if calling constants in that namespace:

$ RUBY_NAMESPACE=1 ruby -ve 'ns = Namespace.new; File.write "ns.rb", "O = Object; OC = Object.constants"; ns.require "./ns"; p Object.equal?(ns::O); p Object.constants == ns::OC; p Object.constants == ns::Object.constants'
ruby 3.5.0dev (2025-05-10T07:50:29Z namespace-on-read-.. bd4f57f96b) +PRISM [x86_64-linux]
ruby: warning: Namespace is experimental, and the behavior may change in the future!
See doc/namespace.md for know issues, etc.
true
false
true

I think this is expected behavior for Namespace so not sure it's worth reopening.

It does mean that with Namespace even though two objects (specifically core classes and modules) might be equal they are actually different when looked at in different namespaces.

[Eregon (Benoit Daloze)]

• Related to Bug #21322: Namespaces and builtin classes as arguments and return values added

[Eregon (Benoit Daloze)]

Seems very related to https://bugs.ruby-lang.org/issues/21322 so let's keep this one closed and discuss there.