[DialogTitle] Test scenario where children are a string

[agamrafaeli]

• PR has tests / docs demo, and is linted.
• Commit and PR titles begin with [ComponentName], and are in imperative form: "[Component] Fix leaky abstraction".
• Description explains the issue / use-case resolved, and auto-closes the related issue(s) (http://tr.im/vFqem).

[oliviertassinari]

I'm gonna accept that PR as the work has already been done. But I don't think it's useful to test the string variation. Thanks.

[agamrafaeli]

Testing the string variation is important on a large scale project because you could get an exception there and never know it.

[oliviertassinari]

@agamrafaeli Sure, but from my perspective, it feels like asserting 4 === 4 after asserting 3 === 3. That looks redundant. I might be wrong though.

[mikew]

Not sure if it's from this PR, but there's different behaviour when using interpolated strings:

<DialogTitle>Edit The Thing</DialogTitle>

yields ...

however ...

<DialogTitle>Edit The {'Thing'}</DialogTitle>

yields ...

It might be necessary to check if children is an array of strings.

[oliviertassinari]

@mikew No link with this PR. Still, that diff is coming from that logic:

      {typeof children === 'string' ? (
        <Text type="title">
          {children}
        </Text>
      ) : children}

I think that it would be better to change the switch logic by something more deterministic, like having a disableText property. Do you want to submit a PR?

[mikew]

Not sure what you mean, I was imagining

areAnyChildrenStrings = Array.isArray(children)
  ? children.filter(x => typeof x === 'string').length > 0
  : typeof x === 'string'

And then the statement you copied above, but branching off of areAnyChildrenStrings

[agamrafaeli]

Thanks @oliviertassinari