rado98 Contributor

May I suggest leaving the post as unsolved in case someone else can come up with a beastmode solution. I would like a beastmode solution to the question too.

I added the request to the Ideas exchange https://dojo.domo.com/t5/Ideas-Exchange/Business-in-a-Box-further-customization/ba-p/34379

Very good question. The not so good way I was forced to do it in the past was using ETL, group by, then aggregate by last value. Not a very stylish solution but worked. Hopefully someone has a better solution

I would suggest starting with the tutorials (Dataflows and Group by Function) then but I will give you info on the function you need to use. The function is Group By 1 Coiumn that identifies the grouping is Item_ID 2 Name fo the new aggregated column can be Available Stock 3 Column to aggegate the fill the new column is…

A way to go around the issue, but nto solve it as such, would be to combine the two warehouses using ETL then either create a separate DataSet or append it to the original and calling it Warehouse A+B in the Warehosue column. Not as clean as the ideal but would work in principle.

Check out this thred for a couple of options https://dojo.domo.com/t5/Beast-Mode-ETL-Dataflow/Adding-Two-Rows/m-p/34211#M5365

I'll tell you how I do it, in the case it applies or would work for you. I hate the idea of appending as the thought of having to worry about data not bieng pushed through and spending hours trying to wokr out what happends plus fixing the gap is just to much. What I do is to create a historical and a current part of the…

The data refreshes any time an email is received

Sorry I should have been more specific. In ETL it is part of the Rank & Window Check out the way to use it in the knowledge base

Hi Try the "Lead" function in ETL a combination of that plus an addition, repeated 5 times will definetely give you what you need (provided that your dataset is indeed that simple. Cheers

I think you are right (specially about side by side graphs), however you should probably post this in ideas exchange or something like to hopefully get things changed/improved

The running total optino should be there, see bellow. If I had to do percent increased and so on I would try something that would use in part the method in this link https://dojo.domo.com/t5/Card-Building/Percent-change-WoW-MoM-etc-as-a-trending-line-in-card/m-p/22965#M2453

Rather than a beastmode you could use the running total option in the card builder (General Chart Properties). That is the quickest, most stream line way. Depending on the card type and fields used it might not be available. If you need to do it via ETL it becomes a tedious 10 step process. I saw a tutorial somewhere but…

The way I have done similar things in the past is Do a join on hte first column Do a join on the second column as a separate step Depending on your data you might even need to filter out all the row form the first join and append them on a later step.

The other key differences I see worth noting are: Workbench requires your PC to do he crunching Workbench does not have a 10000 duplicates limit when doing lookups / joins Workbench is quite clunky in general

I was contacted by support. The Watch File feature does not yet work with network files.

How abouty something like this (using your example) sum (Case... 'Coke'.... `unit sales` ) / sum(`unit sales`) This should allow to filter out States. BeastModes cannot use agregate data with non-agreagated data (as far as i understand), which is a royal pita. 

Hi We did not end up using the function but our Domo support person stated that they could set up something that would allow us to basically email files to Domo daily and those would append automatically into a Dataset. Ask your rep, I would say.

How about something like this For the February scenario above, Create a MONTH= Month('Date') beast mode Select a 3 year time period Add the date field to top right of a Sumo Pivoting table. Set the Date grain to month Set a filter for the MONTH column to February only It should only display Feb 15, 16 and 17 

The excel plug in will do this too. Leaving this as a reference for any future searchers.

The issue was fixed in the latest version after I brought it up with support.

I think you cannot combine aggregated data with non-agreagated data in a beastmode. i.e. Min(DateAlarm) looks a several rows but "Installed Date" look at only 1 row. There is probably a more techincal way to describe this but regardless its a PITA and very inconvenient. Hopefully I am wrong though.

Correct the source and then reupload/replace the incorrect dataset. Transfer cards to the new dataset if need be.

Thanks for the reply. I will wait until WB 4.5 is properly released and try again.

How about something like this in MagicETL Select Columns from the DataInput LIST1 only Select Columns from the DataInput LIST2 only, as a separate stream InnerJoin SelectColumns1 and SelectColums2. This should output a list of only the repeated values. Add the constant "yes" to Join1 Now join again the DataInput with Join1…

It shoud be pretty straight forward: Col1 - Col2 Unless, the columns are not numbers, there is a - / * or other symbols in the column name, or there are digits in the name that are not next to a letter and so on...

Try the "Publish" function. Search for it on the help centre.

Yes, workbench uploads work the same as all other types of DataSets.

Aternatevely, I think you can encript/password protect whole columns if uploading the data usign the workbench. I have not tried it myself though.

I thought I had replied to this already, apologies for the delay. All our files are on a shared server, moving them is not possible nor would I want to if I could. I'll just have to wait for the fix.