From @ee7 in https://github.com/exercism/problem-specifications/pull/1703#pullrequestreview-507842858
But two questions:
Do we have an official ordering that the keys are supposed to be in?
If so, should we enforce the key order during CI? Maybe it helps with readability when adding tests. The ordering seemed pretty consistent before, but there will be nearly double the number of keys now.
I notice that the
reimplementsin this PR is afteruuid, but when I made my recent PRs I instead ordered the keys as I saw them in #1674, that is:{ "uuid": "82d32c2e-07b5-42d9-9b1c-19af72bae860", "description": "two times one is two", "comments": ["Expected value is changed to 2"], "reimplements": "e46c542b-31fc-4506-bcae-6b62b3268537", "property": "twice", "input": { "number": 1 }, "expected": 2 }
I think the order for virtually all of the existing canonical data is:
I'm fine with keeping this as is. Maybe we should add the "reimplements" key directly after the "uuid" key?
Maybe we should add the "reimplements" key directly after the "uuid" key?
I'd like that because it keeps the unreadable part away from the rest
@ErikSchierboom And "scenarios"? In #1674 it was written last in:
{
"uuid": "25a8463d-560f-4a42-9be7-b79d00cd28a4",
"description": "64",
"property": "square",
"input": {
"square": 64
},
"expected": 9223372036854775808,
"scenarios": ["big-integers"]
}
Ah, yes, forgot about that 馃う Scenarios is basically metadata, so it would perhaps make sense to group it with the other metadata. Maybe:
or
I think I have a slight preference for the latter.
I prefer later but not a hard preference.
I prefer the latter too.
Okay, lets go the following then:
Wouldn't this be a change to configlet fmt?
It would, but we're right in the middle of migrating configlet which could take a while.
Most helpful comment
Okay, lets go the following then: