expose {number,time,utc}Interval #2075
Conversation
|
The negative intervals encoding fractional values is something that might be surprising—but I understand if we want to gloss over that. |
It took me a while to interpret this statement but you’re referring to |
|
yes! |
|
Okay, added more documentation, please approve. |
| Plot.numberInterval(2) | ||
| ``` | ||
|
|
||
| Given a number *period*, returns a corresponding range interval implementation. If *period* is a negative number, the resulting interval uses 1 / -*period*; this allows more precise results when *period* is a negative integer. The returned interval implements the *interval*.range, *interval*.floor, and *interval*.offset methods. |
There was a problem hiding this comment.
| Given a number *period*, returns a corresponding range interval implementation. If *period* is a negative number, the resulting interval uses 1 / -*period*; this allows more precise results when *period* is a negative integer. The returned interval implements the *interval*.range, *interval*.floor, and *interval*.offset methods. | |
| Given a number *period*, returns a corresponding range interval implementation. If *period* is a negative number, the resulting interval uses 1 / -*period*; this allows more precise results. The returned interval implements the *interval*.range, *interval*.floor, and *interval*.offset methods. |
remove repeat — or did you mean "when period is a fractional value"?
There was a problem hiding this comment.
No, that’s what I meant, and no, I’m not repeating myself. There are two statements here:
- When period is a negative number, the resulting interval uses 1 / -period, and
- When period is a negative integer, the results are more precise as a result of (1).
If period is a negative number that is not a negative integer, it still works, but there’s not really any reason to do it. So these are saying two separate things.
There was a problem hiding this comment.
(I’m not documenting that a period of 0.5 is automatically promoted to -2, but I feel like that’s an implementation detail that doesn’t need to be documented.)
| /** | ||
| * Given a number *period*, returns a corresponding numeric range interval. If | ||
| * *period* is a negative number, the returned interval uses 1 / -*period*, | ||
| * allowing greater precision when *period* is a negative integer. |
There was a problem hiding this comment.
| * allowing greater precision when *period* is a negative integer. | |
| * allowing greater precision. |
same suggestion/question
Fixes #1994.