Hacker News new | past | comments | ask | show | jobs | submit login

I would hate spreadsheets a lot less if they were testable.

Or didn’t “help” by silent type conversions that skew analyses (which you’ll never notice because, uh, no tests) https://www.sciencemag.org/news/2016/08/one-five-genetics-pa...




Finally, thank you!

Tests, destribution and versioning are my main concerns. I like where the people at stensila are going: https://stenci.la/blog/humane-sheets/


What in spreadsheets is not testable? It's trivial to write a test that opens the spreadsheet in excel and tests any formula calculations via Excel Automation API.


and where do i store my tests? how do i version them, and the data? can i put it in travis?

how many people ever even try?


You store them in the same place where you place other tests. It's just code. You version the data as you version any other test data. Never used travis myself, I guess it's some kind of CI.

Many people do this. It's perfectly normal.

Of course if you don't even use Excel in the first place, I don't know why you're asking.




Consider applying for YC's Fall 2025 batch! Applications are open till Aug 4

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: