jaeW_at_Onyx Coach

The webinar recording is live! We put a ton of Learnings and Resources here. Video Tutorial https://www.youtube.com/watch?v=15nXTopnRcI

I did a python implementation of solving this here: https://www.youtube.com/watch?v=3ZwlzOlRBbA

@JenniferGibson , @MichelleH used Redshift not Magic to get the job done. Beast Modes run on Adrenaline (Domo's database layer), Magic runs in a different environment. Even though many of the functions use similar syntax there is not always 1:1 parity in implemented functions, AND they are not processed in the same space.…

@LiliRestrepo all beta features are available in domo-dojo.domo.com. you don't have to die of Fomo.

@ST_-Superman-_ on paper i think i agree with you but i'm not sure what you want the end experience to be. when you choose a value with the Variable Picker, you want the value of the Variable Picker represented on the Axis? for stuff like this might be helpful to post a quick video of what you mean.

@RobynLinden like ... just in the preview? or as an output dataset? would you want it as part of the execution details page? in the immediate term, I just always tell people to use a GROUP BY or REMOVE DUPLICATES tile before they do a JOIN and that solves any row growth problems... unless there's a NULL in the column.…

@NoahFinberg I LOVE THIS!!!

@andersson_morales19 i think you need CASE and aggregation. so something like CASE WHEN count(column) + count(other_colum) <= 3 AND count(column_3) + count(column_7) <= 3 THEN ... WHEN count(column) + count(other_colum) <= 3 AND count(column_3) + count(column_7) <= 3 THEN ... END

@FXM32 Domo does retain dataset history in Vault, but that's supposed to be a recovery tool not dataset snapshots that end users can flip between. Given that your "freezing mechanism" should tightly align with business logic, it makes sense that you should apply that business logic as an artificat in your dataset that you…

@andersson_morales19 post a sample of your dataset please.

just add a column date_is_not_null and use a CASE statement. CASE when date is null then 0 else 1 end Kimball would say that all dimensional columns should never contain a NULL. Can you fill it?

@SaloniShah post a picture of what you're currently doing / strugglign with.

for the original use case using a formula tile and COALESCE() will be a much cleaner (and recycle-able) design pattern in other parts of the platform. I particularly enjoy the fact that COALESCE is a SQL standard and not bespoke to MAGIC ETL. My challenge with setting data types in the INPUT dataset is last time i looked…

@SaloniShah is the data actually duplicated in your dataset? If so that's probably something to be concerned about if it's not intentional. Take a look at the connector that's bringing in the data, is it set to APPEND or REPLACE? Is the data actually duplicated or are just a few rows duplicates? If you run the connector…

@damen to be more specific, you should (probably) impelment the metric @RobSomers is describing as a Beast Mode in Analyzer. Typically we recommend against preaggregating datasets in ETL if we don't have to.

@trafalger I do wonder if you add the index in the UI if the data is indexed as it's being loaded into the table, or if if the index is added after CREATE TABLE AS SELECT ... So i i guess if you wanted to be REALLY fancy you would CREATE TABLE, ALTER TABLE and then load data... but then the table doesn't appear in the UI.

sounds like your question is whether it's possible to use "platform embed + buzz" i don't know the answer,but maybe drop a support ticket. platform embed is a relatively new feature so it'd be unsurprising if the documentation hadn't been updated.

@Jones01 if a user locks their dashboard, then only admins or owners can edit it.

at the domo user conference @MichelleH showed how she used redshift + geometry functions to get that done. https://docs.aws.amazon.com/redshift/latest/dg/ST_Point-function.html try out your trig functions in beast modes, some of them aren't documented.

my approach would be to try to take a timestamp and date_format( hh:mm) that would always round down. to address round up, i'd use a CASE statement so if second() is 30 then hh:mm + 1 minute.

@Sam_Donabedian some functions are available but not represented in the UI. Did you try it? I would be surprised because beast modes are processed in the same engine as dataset views (AFAIK).

yes the "variables beta" will be going live soon and can address this issue. you'll be able to set up a varaible with enumerated values (ex, "full_price" and "mark_down") then you could write a case statement CASE when <variable> = 'full_price' then <col_name> else <other_col_name> END

@Jones01 , "Domo Everywhere" is an umbrella of tools... what specifically are you looking for?

consider adding a "is_closed_period" column. that way you can filter on transactions that are part of the closed period and functionally "freeze" the data. alternatively, if you label transactions as belonging to the nth period (1,2,3,4th etc.) you could filter on "all activity from the 3 closed period or earlier"

it's unclear what your output from (presumably) magic etl scripting tiles looks like. can you send intermediate screenshots?

i am lead to understand that this is something Domo Product is actively looking at developing.

@TC111 why? most users should not have edit rights to the card... there are apis that will allow you to loop over all the cards in an instance, and surely enable a lock, but in most cases if you're managing roles and groups appropriately, most users should not be able to edit content.

@ST_-Superman-_ and @datamatt , typically for anything like headcount data, you'd want data the granularity of one row per client_id per facility per day. if you don't, imagine i have a customer who was admitted last month and stays for 3 months. in my report, i would only show them on the first day they were admitted. you…

@tobypenn most beast mode functions just use MySQL syntax

@Godiepi that's a pretty obscure use case that you'd expect full-on row duplication. you could use MD5() hashing to concatenate all the values together and then use a RowNumber + rank_window tile then filter where rownumber = 1 this is a classic way of handling identifying slowly changing dimension rows.