Error when providing axis tickformat, but dataset is empty

I could not find this anywhere in reported/open/closed problems, so here goes.
I'm dynamically fetching my dataset for plotly - and sometimes there is no data - a normal condition for my app.
When providing a timestamp-like xaxis tickformat, an error will occur when the dataset is empty.

  Plotly.newPlot(this.chartTarget.id, {
    data: [
      { name: "Spot", type: "scatter", mode: 'lines', x: [], y: [], line: {color: 'black'} },
    ],
    layout: {
      title: "Blah",
      hovermode: "closest",
      hoverlabel: { bgcolor: "#FFF" },
      legend: { orientation: 'h', y: -0.3 },
      xaxis: { tickformat: "%H:%M", title: {text: "Time", automargin: true, standoff: 0}, zeroline: false },
      yaxis: { title: "YAxis" }
    },
    config: { responsive: true }
  });

When the dataset is empty alike above (and only when empty), a JS error will be thrown with a trace that looks like:

console.trace() WARN: encountered bad format: "%H:%M" plotly-finance.js:27425:18
warn plotly-finance.js:27425
warnBadFormat plotly-finance.js:25902
numberFormat plotly-finance.js:50110
numFormat plotly-finance.js:40235
formatLinear plotly-finance.js:40138
tickText plotly-finance.js:39918
setTickLabel plotly-finance.js:39477
calcTicks plotly-finance.js:39509
drawOne plotly-finance.js:40613
draw plotly-finance.js:40541
syncOrAsync plotly-finance.js:26272
draw plotly-finance.js:40534
drawAxes plotly-finance.js:31315
syncOrAsync plotly-finance.js:26272
_doPlot plotly-finance.js:31340
newPlot plotly-finance.js:31501

It seems like the format handler should be able to detect this situation and avoid the error when there's actually nothing to format in the first place.

Thanks for consideration and hopeful for a resolution within plotly itself for this potentially-expected condition.

Reproduce the supplied Plotly.newPlot call with empty x and y arrays and xaxis tickformat "%H:%M". Start from the newPlot call and the format-related stack entries in plotly-finance.js, then confirm that an empty dataset no longer produces the bad-format error.