I'm developing a data migration engine to move data from an older system to a newer one. It's broken down into three sections.
First, obtaining data from earlier tables.
Second. Prepare the request body for API post-processing processes to insert into new tables.
And finally. Which design pattern should I use in this case?
I was considering using the building pattern.
In the next phase, I'll make additional API calls to insert data into new tables. This has some structure to it. If an API request fails in the middle, all prior API calls must be redone. some sort of rewinding mechanism What is the best way for me to accomplish this?