Feature Request for visualization setting in Kibana

[IrdmEngr]

Apologies if any incorrectness about this post, it's my first time on Github.

Kibana version:
7.5.2

Elasticsearch version:
7.5.2

Describe the feature:
Add toggle to turn on/off connecting of data endpoints when missing values during breaks in hits.

Describe a specific use case for the feature:
With line visualizations, if you have a break in your data that picks back up at a later point, the visualization will connect the endpoints even though there are no values/aggregation data between them. Use case - you have three aggregation points over a time span, lets say 1m between each, then you have no values/hits in your ES for the next 3 minutes, then you have data for the next 3m after that. Currently, when turning on connecting lines, you will see the first 3 lines connected as expected, then a long connection line that connects the 3rd point to the 4th point (interpolated values I guess?) then the last 3 lines are also subsequently connected as expected. The request would be to add a toggle to turn on/off the connecting of endpoints when no data. In this case, you would see no interpolated line connecting the 3rd and 4th datapoints since there is no aggregated data in those bucket timeframes.

Attachment is sample of how Excel has the ability to handle this when graphing.

The specific request is to help handle performance of various items that come in and out of visibility in the dataset. if you sub-bucket your aggregations for various contributing items, in my case 20 different items contributing performance that come in and out of data visibility, the current method of connecting all the endpoints makes it look like a big jumbled mess of lines.

[IrdmEngr]

I don't know if this is a bug or what, but it appears in a different visualization that this somewhat works as I'm requesting? Not sure what's going on here since I've compared the settings between the two and they're identical.

[elasticmachine]

Pinging @elastic/kibana-app (Team:KibanaApp)

[timroes]

Hi,

thanks for the detailed explanation. This issue is already tracked via #17717 to allow the user to configure how they want their dataset to be filled in case of null values on the x-axis.

Cheers,
Tim

[IrdmEngr]

Tim-
Thanks for the quazi-update. However, the bug/feature you reference doesn't seem to have any update since October/November - any status on it? I only ask because we had to roll back to a previous version due to some specifics of our operational use case.

[timroes]

This is filed as an issue, but we don't have any estimated timeline yet on it.