In response to https://twitter.com/konstantinhaase/status/558308931781795840

output

Using `an_error_other_than` as an argument to `raise_error`
  failing (FAILED - 1)
  passing

Failures:

  1) Using `an_error_other_than` as an argument to `raise_error` failing
     Failure/Error: expect {
       expected an error other than SyntaxError, got #<SyntaxError: SyntaxError> with backtrace:
         # ./spec/raise_error_spec.rb:12:in `block (3 levels) in <top (required)>'
         # ./spec/raise_error_spec.rb:11:in `block (2 levels) in <top (required)>'
     # ./spec/raise_error_spec.rb:11:in `block (2 levels) in <top (required)>'

Finished in 0.00248 seconds (files took 0.11564 seconds to load)
2 examples, 1 failure

Failed examples:

rspec ./spec/raise_error_spec.rb:10 # Using `an_error_other_than` as an argument to `raise_error` failing

raise_error_spec.rb

RSpec::Matchers.define_negated_matcher :an_error_other_than, :an_instance_of

RSpec.describe "Using `an_error_other_than` as an argument to `raise_error`" do
  example "passing" do
    expect {
      raise ArgumentError
    }.to raise_error(an_error_other_than(SyntaxError))
  end

  example "failing" do
    expect {
      raise SyntaxError
    }.to raise_error(an_error_other_than(SyntaxError))
  end
end

Is there a reason you can't write an expectation about what will happen (e.g. that a specific error is raised) rather than what won't happen?

Yes, it's raising a GH::Error but should not raise a GH::TokenInvalid, which is a subclass of GH::Error so just checking for that wouldn't really work.