Style/EmptyLiteral and frozen_string_literal generate broken code

[adfoster-r7]

---

Expected behavior

Example made up program showing the issue:

result = String.new
amount = gets.chomp.to_i
amount.times do
  result << "hello world\n"
end

puts result

Usage:

ruby foo.rb
2
hello world
hello world

Actual behavior

The Style/EmptyLiteral rule combined with the frozen_string_literal generates an invalid program:

# frozen_string_literal: true

result = ''
amount = gets.chomp.to_i
amount.times do
  result << "hello world\n"
end

puts result

Program error:

$ ruby foo.rb 
5
Traceback (most recent call last):
	2: from foo.rb:5:in `<main>'
	1: from foo.rb:5:in `times'
foo.rb:6:in `block in <main>': can't modify frozen String: "" (FrozenError)

Example rules:

foo.rb:1:1: C: [Corrected] Style/FrozenStringLiteralComment: Missing frozen string literal comment.
result = String.new
^
foo.rb:1:10: C: [Corrected] Style/EmptyLiteral: Use string literal '' instead of String.new.
result = String.new
         ^^^^^^^^^^
foo.rb:2:1: C: [Corrected] Layout/EmptyLineAfterMagicComment: Add an empty line after magic comments.
result = ''
^

Interestingly this problem only arises if the file doesn't already have the frozen_string_literal rule, so having the frozen string literal comment beforehand doesn't actually break the program as the Style/EmptyLiteral rule doesn't run.

# frozen_string_literal: true

result = String.new
amount = gets.chomp.to_i
amount.times do
  result << "hello world\n"
end

puts result

Steps to reproduce the problem

Create a file with the example text above, and run rubocop -A foo.rb

RuboCop version

$ rubocop -V
warning: parser/current is loading parser/ruby27, which recognizes
warning: 2.7.3-compliant syntax, but you are running 2.7.2.
warning: please see https://github.com/whitequark/parser#compatibility-with-ruby-mri.
1.12.1 (using Parser 3.0.1.0, rubocop-ast 1.4.1, running on ruby 2.7.2 x86_64-darwin19)

[adfoster-r7]

The current Style/EmptyLiteral rule is great for fixing uses of Hash.new/Array.new into their literal vaules, but it would be great if the rule allowed more granular configuration value for the string literal functionality to be disabled. Is that something that makes sense to send a pull request for?

[dvandersluis]

@adfoster-r7 thanks, I opened a PR to fix this.