features/docs/getting_started.feature:7 fails on Ruby 2.4.

[voxik]

Summary

The test case features/docs/getting_started.feature:7 fails on Ruby 2.4:

Expected Behavior

The test should pass.

Current Behavior

expected "No such file or directory @ rb_sysopen - features. You can use `cucumber --init` to get started.\n" to include "No such file or directory - features. You can use `cucumber --init` to get started."
Diff:
@@ -1,2 +1,2 @@
-No such file or directory - features. You can use `cucumber --init` to get started.
+No such file or directory @ rb_sysopen - features. You can use `cucumber --init` to get started.
 (RSpec::Expectations::ExpectationNotMetError)
features/docs/getting_started.feature:10:in `Then it should fail with:'

Failing Scenarios:
cucumber features/docs/getting_started.feature:7 # Scenario: Run Cucumber in an empty directory

Possible Solution

This obviously fails due to https://github.com/cucumber/cucumber-ruby/blob/master/lib/cucumber/runtime.rb#L34 . Isn't it about the time to drop the https://github.com/cucumber/cucumber-ruby/blob/master/lib/cucumber/runtime.rb#L15 ? All currently supported Ruby prints this and I don't think it is going to change soon.

[voxik]

Other option is to include it unconditionally ....

[mattwynne]

Thanks for looking into this @voxik!

I'd definitely prefer if we can avoid showing the @ rb_sysopen noise to our users.

I think it would be nicer to stop subclassing the FileException and just raise our own error with our own message, and get rid of all of that cruft. Would you like to send a PR for that?

[lock]

This thread has been automatically locked since there has not been any recent activity after it was closed. Please open a new issue for related bugs.