Fix #321

[hiiwave]

This is to fix #321

[hiiwave]

The tests run well under my environment. I have no idea yet why the CI failed here. (help wanted)

[davidag]

Hello @hiiwave, let me give you my opinion on these changes:

• I would try to avoid raising WatsonError in cli.py. Just use click.ClickException which is in fact what _watson.WatsonError is (still) in cli.py. See Unify WatsonError catching in cli.py #315
• I'm not sure the check for empty project at the beginning of `Watson.start() is still necessary. Did you try to remove and run the tests?
• Could you add a line in the changelog describing the fix?

I hope it helps!

[hiiwave]

Agree. Thanks for the suggestions. Changes applied.

Could you add a line in the changelog describing the fix?

Sure. Is this a common practice for PR here? As I didn't do this in #312, #318 and #320 .

[davidag]

Agree. Thanks for the suggestions. Changes applied.

@jmkerr do you have any clue of why test_report_current is failing with the changes in this PR? Thanks! ☺️

Could you add a line in the changelog describing the fix?

Sure. Is this a common practice for PR here? As I didn't do this in #312, #318 and #320 .

As far as I have seen, it is common practice here.

[jmkerr]

@jmkerr do you have any clue of why test_report_current is failing with the changes in this PR? Thanks! ☺️

This is completely unrelated to the content of the commit, but possibly related to the time the test was run between 00:00 and 01:00 UTC. I'll take a look at the problem and open another issue. (I'm sorry.)

[davidag]

This is completely unrelated to the content of the commit, but possibly related to the time the test was run between 00:00 and 01:00 UTC. I'll take a look at the problem and open another issue. (I'm sorry.)

@jmkerr thank you for your help! If there's some way we can help you out, just say it! 👍