You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I don't see us automatically keeping all these categories around when you hide all the traces, because there is in fact no data displayed, so the category axis has not seen any categories. Consider the case where two traces have different categories in them. Hide one of them and you wouldn't, in general, want to continue to include its categories in the chart - and anyway we would autorange them out of the viewport unless they were intermingled with the categories from the remaining trace.
Consider the case where two traces have different categories in them. Hide one of them and you wouldn't, in general, want to continue to include its categories in the chart
So I guess the reason for linear type axes not to leave out their ticks is because you can't introduce "new types" of ticks there by definition?
It's just kind of confusing by a UI/UX-standpoint to experience such behaviour of complete tick-disappearance only for this case. It would be more consistent to leave out ticks if all traces are invisible, regardless of type, then IMHO.
Here an example from the plotly docs with no specific axis type:
https://codepen.io/plotly/pen/mJYVJd
Here's the same example with axis type set to 'category':
https://codepen.io/anon/pen/mzarWR
When you filter all traces by clicking on the legend you will see that category ticks will disappear.
Current workaround is to leave an invisible trace with
showlegend: false
.The text was updated successfully, but these errors were encountered: