So, from the @Shevy solution I came up with this idea to solve my challenge.
Since Magic ETL doesn’t have the concept of NOW(), I can create it with a MSSQL
SELECT GETDATE() AS CurrentDateTime, '1' AS 'CountofCurrentDateTime'
1/26/2016 2:09 PM
This produces a one record table, I add a ‘Power of One’ field to the table, this will be for matching. If you are not familiar with the 'Power of One', ask Thomas Tobin from SFDC or Steve Molis in the SFDC community.
I now have a field that can approximate the Now() function in the Magic ELT by refreshing the dataset every 15 miutes.
Now I created a new ETL to test my idea,I add a Power of One Field (CountofSampledata) to the SampleData' dataset I want to age by days. I use a Join matching CountofCurrentDateTime & CountofSampledata to add the ‘CurrentDateTime’ stamp to the ‘SampleData’ table. Now, I added a ‘Date Operations’ action that calculates the ‘Days until date’ CurrentDateTime’ minus ‘SampleData.DateTime’. This produces a number value that can be filters.
By doing this with a seperate table it allows me to update the time stamp every time we run the ETL and I am able to use it across mutile dataset. It does not require the main data source to be refreshed to get the datetime stamp.
Thank you @Shevy for the pointing me in the right direction.
Can anybody help @MReznick with their question?Thanks!
A general design theme of Domo is to appropriately manipulate data at the latest stage possible. This generally requires doing less work upfront in the ETL process and allows greater flexibility at the actual card-building level.
Currently, there doesn't seem to be an easy way to filter data in Magic ETL, but it obviously is easy at the card level. You could also use a Beast Mode to create a filter based on the last 30 days that could be used in a card.
Hopefully that helps; but feel free to expound upon your use case here and we can brainstorm other solutions.
@MReznick, did any of the of the above replies help answer your question?