I am developing a SQL app -- bunch of procs and tables called from an SSIS package. I do not have much experience in QA. The guy doing the QA is using a method I am just not too confident about and feel it is causing some issues.
He is passing one or two rows though the app at a time, testing one scenario, then the next, then reporting a bug if it doesn't work. He sends me the bug, I fix it then he runs another record or two through, never resubmitting a scenario once the bug is fixed.
This seems like a real whack a mole method, since changes I make for Row5 could impact Row1. The way I unit test is I create a data set, 50 rows or something, based on the business req - I try to beat it up, run every possible variation and combination I can think of. The results are my baseline. Every time I make a change I run the same data set, compare it to the baseline.
Any thoughts? I'm a dev, I get bugs from QA, never watched the process. Thanks!