I don’t think I've had this much fun since I was at the ATDD pre-conference before Agile 2010 (?) in Chicago.
I'm writing Tables, to express some of my examples. I like my tables to have an extra column explaining why the specific example is important.
Here is one example:
The amounts agree exactly at $50.00, so only the dates are in question.
Each pair below is then judged on its dates alone:
| receipt dated | bank posted | verdict | why |
| ------------- | ----------- | -------- | -------------------------- |
| 2024-01-15 | 2024-01-15 | high | posted the day it was paid |
| 2024-01-15 | 2024-01-16 | medium | posted a day late |
| 2024-01-15 | 2024-01-14 | medium | posted a day early |
| 2024-01-15 | 2024-01-18 | low | three days late, the edge |
| 2024-01-15 | 2024-01-12 | low | three days early, the edge |
| 2024-01-15 | 2024-01-19 | no match | a day past the window |
| 2024-01-15 | 2024-01-11 | no match | a day before the window |
Three days either side is the window. A card takes a day or two to post, so a same-day agreement is the only one we call high confidence.
As it currently stands, the step code needs to take the "why" and echo it back to the test runner. So we have a column compared against itself and extra step code.
Maybe we could have a column named: ignore or some other signal that it is human-only.
I don’t think I've had this much fun since I was at the ATDD pre-conference before Agile 2010 (?) in Chicago.
I'm writing Tables, to express some of my examples. I like my tables to have an extra column explaining why the specific example is important.
Here is one example:
As it currently stands, the step code needs to take the "why" and echo it back to the test runner. So we have a column compared against itself and extra step code.
Maybe we could have a column named: ignore or some other signal that it is human-only.