Legend color turns black if specifying specific colors for each bar

[sijmenvos]

For a project I'm using lighter bar colors to display what bars have insufficient datapoints to draw conclusions from it. Unfortunately this results in the legend color being black.

Is there a workaround to fix this problem? If not it would be a nice feature to be able to set the color of the legend manually for each series.

Example:
https://plot.ly/~sijmen/3/

[etpinard]

Looks like the legend item fallback is wrong for bar traces.

Thanks for pointing this out.

[etpinard]

Reproducible codepen: http://codepen.io/etpinard/pen/JbMbXe

[sijmenvos]

Thanks for fixing this!

It seems that the color of the legend when you deselect a serie it is still black. Normally it's a fade out of the color of the serie.

See the codepen: http://codepen.io/etpinard/pen/JbMbXe
And click on a legend item to deselect it.

Regards,
Sijmen

[aanchalvazir]

Hi @etpinard

I am facing a similar issue. I am creating a bar chart in Dash plotly, to which I am assigning a color scale for showing diverging colors (not assigning colors per bar but a assigning a scale which can do this automatically as per a measure for reference). However, the legend for this chart appears black.

The following is a snippet of what I am trying to achieve:

'data':[
go.bar(
x='dates',
y=( df['mean'] - dff['mean'] ),
marker={'colorscale':'Greens', 'color':(dff['mean'])},
showlegend=True,
legendgroup='group2',
yaxis='y3',
)
],

'layout':[
'yaxis3':{
'domain':[0.8, 1],
'rangemode':'nonnegative',
'visible':False,
'mirror':True,
'range':[0,250000],
'scaleratio': 1,
}
]

I do not want to show the continuous color scale itself as shown in : https://plotly.com/python/colorscales/ but the legend itself.
Could you suggest a work around?

Thanks in advance!

[nitram9]

It seems, this problem is occurring again in 2020.

[archmoj]

It seems, this problem is occurring again in 2020.

@nitramwin could you please share a minimal codepen illustrating the bug you noticed?
Thanks!

[nitram9]

@archmoj Thanks for the response. Here it is: https://codepen.io/nitramwin/pen/mdEoxZo

[archmoj]

@archmoj Thanks for the response. Here it is: https://codepen.io/nitramwin/pen/mdEoxZo

Thanks @nitramwin. Tracked in #5285.