roll doesn't handle periodic boundary conditions well

[shoyer]

DataArray.roll() currently rolls both data variables and coordinates:

>>> arr = xr.DataArray(range(4), [('x', range(0, 360, 90))])
>>> arr.roll(x=2)
<xarray.DataArray (x: 4)>
array([2, 3, 0, 1])
Coordinates:
  * x        (x) int64 180 270 0 90

This is sort of makes sense, but the labels are now all non-monotonic, so you can't even plot the data with xarray. In my experience, you probably want coordinate labels that either look like:

1. The unrolled original coordinates: [0, 90, 180, 270]
2. Shifted coordinates: [-180, -90, 0, 90]

It should be easier to accomplish this is in xarray. I currently resort to using roll and manually fixing up coordinates after the fact.

I'm actually not sure if there are any use-cases for the current behavior. Choice (1) would have the virtue of being consistent with shift():

>>> arr.shift(x=2)
<xarray.DataArray (x: 4)>
array([nan, nan,  0.,  1.])
Coordinates:
  * x        (x) int64 0 90 180 270

We could potentially add optional another argument for shifting labels, too, or requiring fixing that up by subtraction.

Note: you might argue that this is overly geoscience specific, and it would be, if this was only for handling a longitude coordinate. But periodic boundary conditions are common in many areas of the physical sciences.

[fujiisoup]

I like the option 1.
I agree that the circular coordinate is used in many physics fields (actually I'm not from geoscience and I also sometimes use the circular coordinate), but one confusing sene I can think of is when using it in radian, (0, 0.5 pi, pi ...).
Manual overwite would be more explicit

Another option could be to add a circular cooredinate class after #1603 is implemented?

[shoyer]

@mathause @rabernat any opinions here? My sense is that the current version of roll() is not very useful for anyone, but I'd like to check before we change it unilaterally...

[ahuang11]

Just wanted to bump this! Would make my hack neater to interpolate across the prime meridian.

[shoyer]

Yes, it would be nice to change this to behavior (1) above.

There are a few options here:

• Just change how roll works (as a breaking change) in xarray v0.11, with the argument that the current behavior is broken/useless. This is easiest if we aren't concerned about backwards compatibility, but unfortunately I'm pretty sure at least some people rely on this, e.g., based this StackOverflow post.
• Deprecate the current behavior, e.g., by inserting a new argument coords=None which by default shifts coordinates (equivalent to coords=True) and raises a warning. In a future version of xarray, switch the default to coords=False.
• Implement the desired behavior in a new function (roll_axis?) and deprecate roll. This is the preferred option for mature projects like NumPy.

I am loathe to loose the name roll so this second option is probably the best choice. Any interest in putting together a pull request?

[ahuang11]

Option 2 sounds good; I'll try putting together a pull request sometime, hopefully within a week!