0NaN-NaN-NaN NaN:NaN:NaN in Tables when Include Time is enabled

[bearloga]

When checking the "Include Time" option in the GROUP BY section of a chart and using the Table visualization, the output doesn't print the actual datetimes.

Expected results

Datetimes based on time range & granularity selected

Actual results

"0NaN-NaN-NaN NaN:NaN:NaN"

Screenshots

Attaching screenshots from both datasource types (Druid Table, classic cube):

How to reproduce the bug

1. Create a new Table chart
2. Adjust time granularity, range, etc.
3. In the GROUP BY section, check "Include Time" option
4. Run query

Environment

• superset version: 0.36.0

Checklist

Make sure these boxes are checked before submitting your issue - thank you!

• I have checked the superset logs for python stacktraces and included it here as text if there are any.
  • Don't have access to the host but hopefully @elukey can provide
• I have reproduced the issue with at least the latest released version of superset.
• I have checked the issue tracker for the same issue and I haven't found one similar.

[issue-label-bot]

Issue-Label Bot is automatically applying the label #bug to this issue, with a confidence of 0.94. Please mark this comment with 👍 or 👎 to give our bot feedback!

Links: app homepage, dashboard and code for this bot.

[durchgedreht]

Hi,
we encounter the same issue, but with the pattern "NaN/NaN/0NaN".

• superset version: 0.36.0
• clickhouse
• native clickhouse alchemy driver

When looking in the query log, ordering per week applies a toMonday() function after a toDatetime(). Result will not be DateTime but Date (have a look in the. "__timestamp" field - no time infos available). I'm not sure where the issue resides, but the rending in HTML shows a zeroed out Datetime with format "yyyy-MM-dd 00:00:00" (first picture at the bottom shows the chrome dev inspect window with data-sort="2020-02-24T00:00:00" ) .

So I believe it's just the rendering that's impacted. Many NaN Date rendering is related to wrong Date input, so maybe columns are mixed up, wrong format pattern applied, index wrong (month values 0-11 vs 1-12).

I can provide more infos if needed. Please advise which data might be helpful.

[durchgedreht]

Additional infos: Chart's CSV export also uses the date from query. No issues exporting the data.

...,2020-04-27, ...

[durchgedreht]

Seems related to
#10097
and is fixed in 0.37 (not yet testet - will try to verify)
#9493

[mistercrunch]

0NaN-NaN-NaN NaN:NaN:NaN - hey hey hey - goodbye!

[allantaylor81]

the same problem ! Did it work out in the end ？ Look forward to your reply ! thanks

[whitefirer]

0.37 has sloved it.

[durchgedreht]

+1
Works from >= 0.37

[bearloga]

We just upgraded to 0.37.2 and I can confirm this problem is gone. Thank you!

[stevensuting]

This still happens in 0.38.0 for columns that are placed in the group by list. This issue describes the same

This is from my instance:

[villebro]

@stevensuting can you add the following information:

• the database type
• the datatype of the "Last Appointment Date" and "Last Feedback Date" columns?

Also, the cell values that are displaying NaN..., are those NULL values?

[stevensuting]

@villebro

• Database Type: Postgres
• Data type of columns Last Appointment Date" and "Last Feedback Date: timestamptz
• Cell value in DB for which Superset is displaying NaN: NULL
• Data type set in Superset implicitly: TIMESTAMP WITH TIME ZONE

[bonedi]

I still have this Issue in version 2.1.2.
I have tried to cast the value to varchar, still I get 0nan-nan... in the chart.

[MilenaSilva10]

I also have this problem, how can I solve it?

[villebro]

@MilenaSilva10 which version are you on, and which database type? This should not be happening on recent versions of Superset.

[MilenaSilva10]

@villebro it is showing version 0.0.0-dev. I saw it happened in August 2023 too, but I think the problem came back
#24696