[GitHub] [superset] michael-s-molina commented on pull request #13802: feat: don't return empty time-series

[GitBox]

michael-s-molina commented on pull request #13802:
URL: https://github.com/apache/superset/pull/13802#issuecomment-808185276


   I think that both zero and null values can or cannot be important for a 
chart depending on context. 
   
   Let's imagine that this chart is the number of daily deaths by Covid:
   
   <img width="814" alt="Screen Shot 2021-03-26 at 9 24 08 AM" 
src="https://user-images.githubusercontent.com/70410625/112631007-1dc77380-8e15-11eb-9632-e9f9cf34b703.png";>
   
   - The first example (gap) could demonstrate how the null values influenced 
the data, clearly indicating to the user that some days could not be processed.
   - The second example (zero) could demonstrate that no deaths occurred on a 
specific day.
   - The third example would be the chart ignoring nulls and zeros.
   
   As we can see the information can be very different. I think this behavior 
should be configurable through filters, calculated metrics, or some feature 
like the one we have in Excel:
   
   <img width="328" alt="Screen Shot 2021-03-26 at 9 33 49 AM" 
src="https://user-images.githubusercontent.com/70410625/112632083-69c6e800-8e16-11eb-99e0-f61719ff63bb.png";>
   
   I think the same principle is applicable to bar charts, area charts, etc 😉 
    


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org



---------------------------------------------------------------------
To unsubscribe, e-mail: notifications-unsubscr...@superset.apache.org
For additional commands, e-mail: notifications-h...@superset.apache.org