Mocha discourages arrow functions in tests

[wkw]

Just getting started with first FJS project. Was working on initial tests this morning and ran into this while trying to increase a test's timeout:

Passing arrow functions to Mocha is discouraged.
Their lexical binding of the this value makes them unable to access the Mocha context, and statements like this.timeout(1000); will not work inside an arrow function.

https://mochajs.org/#arrow-functions

Are there any arguments for sticking with Arrow functions for generated tests?

[daffl]

Not really, it is probably a good idea to change it back to normal functions as they suggest. If you have the time, a pull request changing the test files in https://github.com/feathersjs/generator-feathers/tree/master/generators would be awesome.

[wkw]

OK, I'll take a look through the genies and try to get a suitable PR put together.
Be good to have a little involvement with the project.

[wkw]

@daffl any advice on how I tell Yeoman to use my modified generators for testing the changes? I've only dealt with Yo as a user and not even that often.

[daffl]

In the generator repository run npm link. Then any calls to yo feathers:* will use that linked repository. For those changes you can also just run npm test to see if they are still all passing.

[wkw]

One thing that stumped me after I first updated the generated tests was that none of the non-transform tests were succeeding, all for exit codes not being zero. I eventually added a console.log to show the tmp dir path so I could go in and run the tests directly on the generated app. It was jshint complaining about semicolons. Seems like there should be a way to pipe those jshint failures to the output so it's clearer why the test failed. I'm not sure how that would be accomplished, but would gladly take pointers and work up another PR.

[daffl]

Thanks a lot for the PR! I noticed, too that there wasn't a lot of error reporting when the app test command fails so I added feathersjs-ecosystem/generator-feathers@f006e4d to the tests. Now we should get an error with the full output of the test run.

I'm going to close this issue and will make a new release of the generator shortly.

[wkw]

@daffl I saw your commit earlier for the better error reporting and of course had to try it out. Perfect!

[lock]

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