[TSVB] Stop inserting zeroes for null series

[wylieconlon]

TSVB finds the series names first, and then applies the date histogram after. This leads to cases where a series will be displayed even though there is no data at all for the time bucket that it's showing.

I have created an extreme example where the last value of the series is totally missing, and it will show an empty value instead of zero. In most cases this won't happen, most users will see numbers:

Fixes #90220

Checklist

• Unit or functional tests were updated or added to match the most common scenarios

[elasticmachine]

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

[wylieconlon]

@elasticmachine merge upstream

[stratoula]

@elasticmachine merge upstream

[stratoula]

@wylieconlon thanx for this, it works fine! My only concern is that it also affects Count. So here the count for rpm is 0

and on the metric tab, the chart appears with no value

which may be confusing. I mean count 0 makes sense to me. I would love your thoughts on that.

[alexwizp]

@stratoula @wylieconlon I agree with both of you. We should definitely remove that wrong 0. It confuses me from the beginning of using TSVB. Especially if we work with min/max aggregations and this zero appears. It is difficult to understand this due to lack of data or is it a real calculated value.

Small suggestion, I think from a UI point of view, in such cases it's better to se - instead of empty area.

For Count aggregation I'm ok with both (- , 0)

[wylieconlon]

@stratoula @alexwizp thanks, I agree that this would make more sense. Will update.

[wylieconlon]

@stratoula @alexwizp I fixed the default values and handling of 0 vs null.

[stratoula]

@elasticmachine merge upstream

[kibanamachine]

💚 Build Succeeded

• continuous-integration/kibana-ci/pull-request
• Commit: f63276a
• Storybooks Preview
• Documentation Changes

Metrics [docs]

Async chunks

Total size of all lazy-loaded chunks that will be downloaded as the user navigates the app

id	before	after	diff
visTypeTimeseries	1.7MB	1.7MB	+149.0B

History

• 💔 Build #108250 failed 20288fe
• 💔 Build #108129 failed 2015c63
• 💔 Build #108137 failed d62ef13
• 💚 Build #107191 succeeded fca7a36
• 💚 Build #106872 succeeded c1d386c

To update your PR or re-run it, just comment with:
@elasticmachine merge upstream

[stratoula]

Code LGTM, works as expected, thank you @wylieconlon!

[alexwizp]

LGTM! Tested locally, thank you for that

[poffdeluxe]

Test update looks good to me