And RSpec were also adjusted to the new behavior AFAICT: https://github.com/rspec/rspec/pull/282/commits/1c20fa80772ca7a1ed0512056ce7cd6a94f8e68dvo.x (Vit Ondruch)
Just FTR, not sure how the revert is supposed to look like, but at least Pry was adjusted for the new behavior: https://github.com/pry/pry/pull/2357 It does not benefit from columns, but I suspect the revert would somehow influence...vo.x (Vit Ondruch)
matz (Yukihiro Matsumoto) wrote in #note-8: > I think“Ruby's License” is something described in COPYING, i.e., it is inseparable from BSD-2clause. Historically it was combination with GPL, so some description should be updated. > ... T...vo.x (Vit Ondruch)
vo.x (Vit Ondruch) wrote in #note-10: > And I have not tried what would happen if the array returned by `instance_variables_to_inspect` contained some garbage ... Just tested a few examples: ~~~ $ irb irb(main):001* class Foo i...vo.x (Vit Ondruch)
byroot (Jean Boussier) wrote in #note-9: > I think the better solution here is to pre-define `Kernel#instance_variables_to_inspect` That solves one part. But if this is considered acceptable, could the error message (and/or documenta...vo.x (Vit Ondruch)
In retrospect, is the `instance_variables_to_inspect` design good? Wouldn't it be better if such method behaved as some filter? E.g. it would get e.g. an array of something (maybe list of instance variables?), which could be altered and ...vo.x (Vit Ondruch)
This is essentially what happens: ~~~ $ irb irb(main):001* class Foo irb(main):002* def initialize irb(main):003* @bar = "bar" irb(main):004* end irb(main):005> end => :initialize irb(main):006> Foo.new.inspect => "...vo.x (Vit Ondruch)
To me, it seems that the `VALUE ivars = rb_check_funcall(obj, id_instance_variables_to_inspect, 0, 0);` code at `rb_obj_inspect` assumes that there is returned either `nil` or `Array` of symbols. But with `method_missing` available, it c...vo.x (Vit Ondruch)
> I only checked minimagick, this looks caused by https://bugs.ruby-lang.org/issues/21219 Yep, I think so. I have just tried to add `p composite` on [this](https://github.com/minimagick/minimagick/blob/ddcbfe9632667876be10373e1e012...vo.x (Vit Ondruch)
Testing Ruby 4.0 on Fedora, we observe these errors in [MiniMagick](https://github.com/minimagick/minimagick/) ~~~ 1) MiniMagick::Image#composite yields an optional block Failure/Error: yield composite if block_given? TypeE...vo.x (Vit Ondruch)
vo.x (Vit Ondruch)11:26 AM Ruby Feature #6012: Proc#source_location also return the column