fix(OrdinalScale): fix line graph renders null value incorrectly. clo…

[WindyZ99]

…se #16664

Brief Information

This pull request is in the type of: bug fixing

What does this PR do?

Fix function OrdinalScale.parse returns an unexpected value 0 with null.

Fixed issues

• [Bug] Line graph with Y axis "category" renders null values incorrectly #16664

Details

Before: What was the problem?

When options like this:

options = {
        animation: false,
        xAxis: {
            type: "category",
            splitLine: { show: true },
            data: ["1:00", "2:00", "3:00", "4:00", "5:00"]
        },
        yAxis: {
            type: "category",
            axisLabel: {
            show: true
            },
            showMinLabel: true,
            showMaxLabel: true,
            data: ["A", "B", "C"]
        },
        series: [
            {
            type: "line",
            data: ["null", "A", undefined, "C", null], // with value `null`
            symbol: "circle",
            showSymbol: true,
            symbolSize: 8,
            lineStyle: {
                opacity: 0
            },
            connectNulls: false
            }
        ]
    };

We expect the value of null will not be rendered,
but ECharts renders it like the following:

When OrdinalScale.parse function handle the input value of null,
it returns 0 rather than NaN,
because Math.round(null) will produce 0.
We expect null will behave the same as undefined.

So I add a judgement in OrdinalScale.parse.

parse(val: OrdinalRawValue | OrdinalNumber): OrdinalNumber {
        // Caution: Math.round(null) will return `0` rather than `NaN`
        if (val == null) {
            val = undefined;
        }
        return isString(val)
            ? this._ordinalMeta.getOrdinal(val)
            // val might be float.
            : Math.round(val);
    }

After: How is it fixed in this PR?

[echarts-bot]

Thanks for your contribution!
The community will review it ASAP. In the meanwhile, please checkout the coding standard and Wiki about How to make a pull request.

[pissang]

I think we can return NaN directly here

[WindyZ99]

Yeah, I didn't realize this.

[susiwen8]

• This PR also close Null values with category-based Y-axis don't render correctly #15579

[echarts-bot]

Congratulations! Your PR has been merged. Thanks for your contribution! 👍