[TSVB][Lens] Add "open in lens" functionality for Top N

[VladLasitsa]

Summary

Completes part of #138236.

As part of phasing out TSVB and Visualize all TSVB visulizations should support "open in lens" functionality.
In that PR converter for Top N was added.
Also was added support of the following stuff:

• value count agg
• standard deviation agg
• last value mode

Top N in TSVB:

In Lens:

[elasticmachine]

Pinging @elastic/kibana-vis-editors @elastic/kibana-vis-editors-external (Team:VisEditors)

[VladLasitsa]

@elasticmachine merge upstream

[VladLasitsa]

@elasticmachine merge upstream

[stratoula]

Hey! Thanx for working on this. Some comments and questions from my side:

I wonder why don't we allow the transition for multiple layers? For example:

can be translated to

As long this is not allowed in TSVB, we should not allow the transition to Lens either

The last value mode should be converted to Lens with the Reduced time range setting. Am I right @flash1293 ?

There are some aggs that work on TSVB such as the cumulative sum (due to the date_histogram) but not in Lens. Should we allow this transition? cc @flash1293

In case of percentile ranks split by terms the custom color is not transferred in Lens. So for example this will transition to Lens with the default green color.

[flash1293]

I wonder why don't we allow the transition for multiple layers

Vlad and me discussed this - as the chart looks pretty weird due to the empty slots for the other series, we decided to not convert these for now. However, looking at it again I think it's worth doing the conversion even if it looks a little weird - better than no chart. @VladLasitsa could you change this? Sorry for the back and forth

The last value mode should be converted to Lens with the Reduced time range setting. Am I right

yes exactly! I didn't mention this in the doc. We can also split it out. I'll update the doc

There are some aggs that work on TSVB such as the cumulative sum (due to the date_histogram) but not in Lens. Should we allow this transition?

No, let's block these please.

[stratoula]

Thanx Joe! I had updated the doc some days ago if I recall correctly!

[VladLasitsa]

@elasticmachine merge upstream

[VladLasitsa]

@stratoula, Could you please review again?

[stratoula]

Thanx Vlad! Found one bug!
If I am on TopN last value and then go back to timeseries and click "Navigate to Lens" I see this error

This happens because we also transfer the last value mode which is wrong as the timeseries chart is not compatible with this.

[VladLasitsa]

@elasticmachine merge upstream

[VladLasitsa]

Thanx Vlad! Found one bug! If I am on TopN last value and then go back to timeseries and click "Navigate to Lens" I see this error

This happens because we also transfer the last value mode which is wrong as the timeseries chart is not compatible with this.

Fixed

[flash1293]

Looks pretty good, just found one bug: "Last value" mode is not respected for formula-type dimensions (like standard deviation or filter ratio) - it seems like it's just ignored for these cases.

[VladLasitsa]

Looks pretty good, just found one bug: "Last value" mode is not respected for formula-type dimensions (like standard deviation or filter ratio) - it seems like it's just ignored for these cases.

@flash1293, Could you please re-test?

[flash1293]

Seems like filter ratio is not convertable anymore:

Maybe the check for "missing field" went wrong?

It also still tries to set the reduced time range when on a timeseries chart:

[flash1293]

As discussed in the weekly sync, parent pipeline aggregations should not be convertable for charts other than time series because they are executed along the time axis and there isn't a time axis in Lens:

This conversion is not equivalent, the overall_sum is doing something else in Lens than in TSVB

[flash1293]

LGTM, tested and all cases work as expected. I will follow up on the moving average problem

[ppisljar]

app services changes lgtm

[VladLasitsa]

@elasticmachine merge upstream

[kibana-ci]

💚 Build Succeeded

• Buildkite Build
• Commit: 4e31a29

Metrics [docs]

Module Count

Fewer modules leads to a faster build time

id	before	after	diff
visTypeTimeseries	361	363	+2

Public APIs missing comments

Total count of every public API that lacks a comment. Target amount is 0. Run node scripts/build_api_docs --plugin [yourplugin] --stats comments for more detailed information.

id	before	after	diff
data	2428	2431	+3
visualizations	386	388	+2
total			+5

Async chunks

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

id	before	after	diff
lens	1.2MB	1.2MB	+182.0B
visTypeTimeseries	453.0KB	459.8KB	+6.8KB
visualizations	197.4KB	197.4KB	+88.0B
total			+7.1KB

Page load bundle

Size of the bundles that are downloaded on every page load. Target size is below 100kb

id	before	after	diff
data	429.5KB	429.6KB	+52.0B
visTypeTimeseries	18.7KB	18.9KB	+178.0B
total			+230.0B

Unknown metric groups

API count

id	before	after	diff
data	3114	3117	+3
visualizations	414	416	+2
total			+5

async chunk count

id	before	after	diff
visTypeTimeseries	11	13	+2

History

• 💚 Build #65986 succeeded a328328
• 💚 Build #65961 succeeded 4c97ac3
• 💚 Build #65292 succeeded d32db79
• 💚 Build #64791 succeeded 059c42e
• 💚 Build #64148 succeeded 87559df
• 💚 Build #63843 succeeded f05039f

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

cc @VladLasitsa