new(axis): pass all tick values in tickLabelProps signature

[williaster]

🚀 Enhancements

This updates the signature of @visx/axis's Axis.props.tickLabelProps:

// before
(tickValue, index) => TextProps

// after
(tickValue, index, tickValues) => TextProps

This is consistent with how many d3 accessors work (gives you access to the data point, index, and all data points) as well as matches the signature of Axis.props.tickFormat.

I noticed this as a limitation when using xychart in an internal app (e.g., if you want to right-align the last tick, but don't know how many ticks you have since numTicks is approximate).

@kristw @hshoff

[williaster]

have been updating test('it ...') to it('...') as I've encountered them when updating tests

[coveralls]

Pull Request Test Coverage Report for Build 532545191

• 3 of 3 (100.0%) changed or added relevant lines in 2 files are covered.
• No unchanged relevant lines lost coverage.
• Overall coverage remained the same at 61.749%

---

Totals
Change from base Build 516508911:	0.0%
Covered Lines:	1786
Relevant Lines:	2753

---

💛 - Coveralls

[hshoff]

Does this mean tsc will require folks to update the arguments for every tickLabelProps? guess I'm trying to figure out if a tsc warning/error on this would break builds without everyone going in and updating all of their functions to the new signature

[kristw]

Passing function that takes fewer arguments should be ok.
Passing function that requires more argument is not.

type MyFnType = (a: string) => string;

// OK
const f1: MyFnType = () => 'a';
// OK
const f2: MyFnType = (a: string) => a;
// error
const f3: MyFnType = (a: string, b: string) => 'a';

[williaster]

^yeah I don't think that using the old signature will be problematic (value, index) => TextProps shouldn't throw. One thing that would break is if folks are using TickLabelProps explicitly in their typing:

// error
const fn: TickLabelProps<ValueType> = (value, index) => TextProps

That would throw because the signature doesn't match (XYChart wraps Axis and uses that type, so I had to update the signature on this line)

I'd guess that would be rare. I don't think we've considered type changes like that breaking historically (e.g., bumping d3 types wasn't breaking even though there were maybe breaking changes)

[kristw]

👍