Ability to create a custom aggregation for table summary row

[Josh_Bullough]

Add the option to create a custom aggregation through JavaScript to the summary row dropdown as shown in image.

[trz-justin-dev]

looks like plotly added min/max and first/last aggregation functions within a day. otherwise plotly JS only uses JSON so unlikely. everything else i found uses dash

Additional aggregation functions: range and change

Additional aggregation functions

good request though regardless

[jg80]

Just want to cross-post these two other threads on related topic of additional aggregation functionality/controls for the table:

[thread] New Table Feature Announcements

Not sure if this has been split off into another topic, but there are some oddities and some feature requests in the aggregation functionality that would be nice to resolve: Issue: When aggregating by Average, NULL values are interpreted as 0. While possibly valid if a user intends NULL to be interpreted as 0, but usually NULL is meant to be NULL and calculation of average should ignore these records. In other words, average(2, 3, 0, 7) = 12 / 4 = 3 while average(2, 3, NULL and 7) = 12 / 3 =…

Custom Grouped Aggregation

In a table, when grouping by columns, I can select an aggregation type such as sum, which conveniently gives me a total for that group in the "grouped row". However, if I have a column with custom calculations, ex. column c / column d, the grouped row aggregation is calculated incorrectly, no matter what type of aggregation I select. If I don't select an aggregation, nothing shows up. How do I get the grouped row to show the correct calculated aggregated total when columns are grouped? Ex Table …

[jcmarcano]

I have been working with group tables and found that the use of max and min in all column types would be helpful, and it works with any data type.
My use case needs to aggregate a Boolean column. In my case I need an "and" aggregation, which can be solved with a "min" aggregation (following the javascript standard: false=0, true=1), and a "max" could work as an "or".

[AbbeyHernandez]

Hey everyone,

I was surprised to see 4 different names that start with a "J"!

This has been requested in the past. I've logged this conversation to the ticket internally. Will check-in with any updates. I appreciate you all jumping in to collaborate!

[syardumi]

+1

I need a custom aggregation for calculating a division between 2 aggregated row cells as opposed to an aggregation (average, sum, min, max, etc) that uses the values of the entire column.

[drewski]

+1 for custom aggregation (or even custom values from a separate subtotals/totals query) in grouped summary and table summary rows.

[AbbeyHernandez]

Thank you! I've logged these plus ones!

[alexkwolfe]

+1 for us too! Thanks!

[AbbeyHernandez]

Added to our feature request tracking. Thanks Alex!

[drewski]

Hey @AbbeyHernandez - Does this look like a feature that is going to added soonish? Or on the backburner?

[AbbeyHernandez]

No timeline on this just yet, but I will comment here if anything changes.

[drewski]

Thanks, @AbbeyHernandez .

For additional context, grouping rows is unusable when the table includes averages, as they need to be recomputed from the underlying data to provide an accurate number at a higher level. You can avoid this issue for the summary row at the bottom of a table by not defining a summary for columns that are averages, but it would at least be nice to have the summary excluded from the Grouped row within the table as well, since it just puts bad data into the table in the grouped row.

A quick fix (until what we describe above is available) would be to exclude column aggregations in grouped rows when no summary aggregation is specified for that column (it seems like the column aggregation just defaults to Sum right now anyway).