Works for me
How often happened to you, to a data engineer or data tester, that you just simply couldn’t find the root cause of a data issue? Everything look alright, but whatever testing tool you use, simply shows there’s an error. And suddenly, few hours later, voila here it is! What a relief then.
Let us share with you one more story from our life with CAT.
My data can't be wrong
Customer:
“CAT indicates our data is wrong, but it actually isn’t! Non-English characters are handled in a wrong way – it cripples letters with accents on one side of the comparison. What code page does CAT use?”
CAT Support:
“CAT just simply runs your queries and compares the results, are you sure your data is OK?”
Customer:
“The data is correct, something is wrong with CAT, we need to fix it ASAP… or, wait a minute!”
OK, my data ... is wrong
The case turned out the non-English characters got lost between two stages of customer’s DWH load process. So “Väinö Mäkelä” became “Vaino Makela” somewhere on data processing path. Luckily CAT caught that thanks to a very simple test. No bug in the testing tool, rather in the code it was testing 🙂 But that’s what testers life is about ever since – when hunting for bugs, nothing should be trusted.
It is astonishing how much you can check when comparing results of two very simple SELECT statements:
TODO: CODE DISPLAY
- The data in the DWH is fresh – it contains latest customers.
- The data is complete, the incremental load works and nothing was missed when the load was down for two days.
- The data is correct, no problems with non-English characters.
- Sorting works as expected, so collation is OK.
That’s a lot of feedback, considering you just compared two result sets, isn’t it?
Will you share your stories?
We’d love to hear from you – would you share any funny moments from testing? Please share in comments on our LinkedIn profile.
Thanks and happy testing!
