indent test contexts

[charlierudolph]

Joining the all the test contexts by spaces makes it hard to parse. I've never attempted to read that string as its typically impossible to tell where one title ends and another begins.

becomes

If this will be accepted, I'm happy to work to remove fullTitle() from its other uses but for now just concentrated on this one that effects me.

[coveralls]

Coverage increased (+0.06%) to 88.35% when pulling aa24e82 on charlierudolph:cr-indentContexts into e249434 on mochajs:master.

[ScottFreeCode]

While considering any changes to title reporting we will need to keep in mind compatibility with tools such as karma-mocha and potential future support of https://github.com/js-reporters/js-reporters in addition to any general expectations of parity with other test systems or backwards-compatibility (whether with custom reporters or with anything that might be processing reporter output).

[charlierudolph]

Looking at js-reporters it has the following for a given test:

fullName: Array - array of strings containing the name of the test and the names of all its suites ancestors.

We could deprecate fullTitle and introduce titlePath as is or with a new name.

[boneskull]

@charlierudolph Thanks.

Yes, let's leave fullTitle() as-is, as @ScottFreeCode suggests. I think this change is probably a good one, but it's semver-major.

I'd encourage you to use this in a custom reporter for now, as I'm not sure when the next major will land.

[charlierudolph]

@boneskull sounds good.

I just resolved conflicts but am otherwise curious as to what caused the addition of the pr-needs-work label

[coveralls]

Coverage increased (+0.06%) to 88.456% when pulling 296cb23 on charlierudolph:cr-indentContexts into f20de56 on mochajs:master.

[boneskull]

@charlierudolph it looked to me like there were changes that @ScottFreeCode wanted, but maybe I was incorrect on that. I'll let him comment and/or remove the tag

[ScottFreeCode]

I'm actually not sure I know the stuff I brought up well enough to make any definitive statements about impact to this PR; I just wanted to make sure present compatibility (and in some cases desired future compatibility) with those various things is taken into consideration.

[charlierudolph]

@ScottFreeCode so is there someone else we should have look at this? Is there any path forward for this PR or are we blocked until the present compatibility is determined so we know what actions need to be taken with regard to this PR?

[ScottFreeCode]

I believe the @mochajs/core team is still discussing where to merge semver-major changes into. I'm expecting a new semver-major branch for this purpose, but I don't know if we reached consensus on that.

If there are no other concerns from the team then that should be all we need. (Besides which, we could always revisit the details in between merging this in as-is and releasing the next semver-major.)

[charlierudolph]

So does this PR still need the label pr-needs-work? Based on the current comments I think it needs a different label

[ScottFreeCode]

Sorry, I missed "and/or remove the tag" in boneskull's comment when I read it before; I've gone ahead and removed it.