Bring your best ideas to the AI Use Case Contest! Enter to win 40 hours of expert engineering support and bring your vision to life using the powerful combination of Alteryx + AI. Learn more now, or go straight to the submission form.
Start Free Trial

Alteryx Designer Desktop Discussions

Find answers, ask questions, and share expertise about Alteryx Designer Desktop and Intelligence Suite.

Desktop Designer | Use output of previous run to compare with input of current run

alteryxuser20
5 - Atom

Hello everyone,

I have created a workflow which takes in a list of UNIQUE_IDs, lets call it 'previous_input'

Then the workflow joins with other data, does some logical calculations and creates an output.

Lets call this output as 'previous_output'

Similar terms 'current_input', 'current_output' for current/next run.

 

Within the same workflow, I need to use UNIQUE_IDs from previous_output, to compare with UNIQUE_IDs from current_input and process only the new UNIQUE_ID to give result as current_output.

 

So, if you see, there is an underlying circular reference, because I am using output as an input within the same workflow.

 

Using "Block Until Done" tools, I am able to output the data (previous_output) to a folder. My current_input reading begins only after writing previous_output is complete.

So timing is not a problem.

 

However, when I try to use "Join" tool (Left = current_input, Right = previous_input, joining field = UNIQUE_ID to take Left join output), I get 'Loop in connection tools: cannot execute error'.

I understand the circular reference nature of my situation but is there any way to get around this?

 

Unfortunately, I am not able to share my workflow with the community due to security reasons.

 

If anyone can use mock data and help out with how to overcome such problem, I will truly appreciate it.

 

Thanks in advance!

1 REPLY 1
ChrisTX
16 - Nebula
16 - Nebula

You mentioned "current/next run" and you mentioned "within the same workflow".  So it's not easy to visualize the data flow.

 

Would it be possible to create a simplified mock workflow to show your data stream(s), and where you're getting the error?

 

I'm thinking you may need a batch macro (if the logic is the same for "current/next run"), or a chained app.

 

Chris

Labels
Top Solution Authors