Can we do it? Yes we can!
Last month we shared an interesting use-case in which the Mule Whisperer rose to the challenge.
Having delivered, the users came back with a new use-case. This one had an interesting twist.
We need to create a report using data from two different data sources where there is no direct link between the data.
The Process API should figure out which other Data Sources to add and then build the ‘route’ that links said data sources on the fly.
This posed a particularly interesting challenge. It amounted to the difference between giving the automated process a specific set of directions (e.g. First left, then second right. Then go left at the T-Junction and after one mile make another right. Destination is third house on the left) vs giving the process a starting point and an ending point, then leaving it to figure out how it will map the route between those two points (e.g One field is in this Database. Another field is in that file; you figure out how they might relate to each other and give me the report.
The users said, ‘Can we do it?’
I replied, ‘No we can’t.’
Then I thought about it for a few days. After that, I went back and said, ‘I think there might be a way. Maybe. I’m not sure but I am optimistic. Let’s at least give it a try.’
The users agreed, so I went to the product stable (in this case, MuleSoft’s Anypoint Studio).
I whispered …
… and MuleSoft delivered!
No external code required! It found the routes, it mapped them and it joined the data. The best part for me and my team were those tests where the Process API actually discovered routes between Data Sources that we, the DBA’s and the Enterprise Architects had not spotted and didn’t expect. No, it wasn’t AI — but it felt like it!
The client had set a seemingly insurmountable task. They knew it was a big ask but they asked nonetheless, ‘Can you do it?’
We did.