[ruby-core:97535] [Ruby master Feature#16175] Object#clone(freeze: true) is inconsistent with Object#clone(freeze: false)

Issue #16175 has been updated by jeremyevans0 (Jeremy Evans).


Pull request 2954 was merged early today, so I was able to work on this.  I submitted https://github.com/ruby/ruby/pull/2969 to implement it.

In the recent dev meeting notes, @akr stated `freeze: nil` should be supported, but @matz said he didn't see a need to introduce it.  After the merging of pull request 2954, Object#clone is partially written in Ruby via `kernel.rb`, and it would require more complex code not to support it.  Additionally, `delegate` and `set` would both require more complex code if it was not supported.  Therefore, I implemented support for `freeze: nil`, and made `nil` the default value of the `freeze` keyword argument for `initialize_clone`.

----------------------------------------
Feature #16175: Object#clone(freeze: true) is inconsistent with Object#clone(freeze: false)
https://bugs.ruby-lang.org/issues/16175#change-84696

* Author: zverok (Victor Shepelev)
* Status: Open
* Priority: Normal
----------------------------------------
In #12300, the new keyword `freeze:` was introduced, allowing this:

```ruby
h = {}.freeze
h.clone.frozen?
# => true
h.clone(freeze: false).frozen?
# => false
```
Though, it turns to me that behavior is not symmetric:
```ruby
h = {}
h.frozen?
# => false
h.clone.frozen?
# => false
h.clone(freeze: true).frozen?
# => false -- I expected true here!
```
I wonder, if it is "by design" and should be addressed in docs, or just an implementation inconsistency that can be fixed?

---Files--------------------------------
freeze-true.patch (1.47 KB)
clone-freeze-true-16175.patch (4.42 KB)


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

Unsubscribe: <mailto:[e-mail]?subject=unsubscribe>
<http://lists.ruby-lang.org/cgi-bin/mailman/options/ruby-core>