Hacktoberfest 2026: die Issues, die Maintainer für den Oktober markiert haben – offen und einsteigerfreundlich. Hacktoberfest-Issues durchsuchen

uxarray type dropped during `xr.Dataset + UxDataArray` and `ufunc(xr.DataArray, UxDataset)`

Offen
  • Stimmt etwas nicht?
#1,802 0 Kommentare 0 Reaktionen 0 zugewiesene Personen Auf GitHub ansehen

Maintainer antworten meist innerhalb von 3 Tagen

Dieses Issue hat noch niemand übernommen.

Bewertung

Schwierigkeit
5/5
Geschätzter Aufwand
Über eine Woche
Anfängerfreundlichkeit
35/100
Issue-Typ
Bug
Klarheit
Größtenteils klar
Aktivitätsstatus
Aktiv
Tech-Stack
python
Bereich
data

Rechercherichtung

Run the provided MCVE for the three mixed xarray/uxarray cases and inspect xarray method dispatch together with NumPy array_ufunc handling. Determine the xarray hook needed for these operations to preserve UxDataset, then verify that all listed binary operations and ufunc examples return UxDataset.

Vom Indexierungsmodell aus dem Issue-Text verfasst.

Beschreibung

bug fix upstream
Version

v2026.09.1

How did you install UXarray?

Source

What happened?

Spinoff of #1685 and #1695 to track a few remaining difficult-to-solve cases, which should now be considered out of scope for those issues. Results from binary operations with one xarray object and one uxarray object, when one is a Dataset and the other is a DataArray, currently do not always return a UxDataset (but they should).

(Workaround for any users encountering this bug: explicitly cast objects to uxarray types before or after performing the operation.)

These cases are:

  1. xr.Dataset * UxDataArray, where * can be any binary operation (e.g. multiplication, addition, subtraction)
  2. ufunc(xr.Dataset, UxDataArray), where ufunc can be any numpy binary ufunc (e.g. np.minimum, np.divmod)
  3. ufunc(xr.DataArray, UxDataset), where ufunc can be any numpy binary ufunc (e.g. np.minimum, np.divmod)

In the docstring of test_nondestructive_binary_ops_and_binary_ufuncs_output_types() in #1767, these cases map to:

  1. case (K)
  2. also case (K)
  3. case (J)

All of these issues cannot be fixed in uxarray directly. Rather, to get these properly fixed would require editing xarray directly to, at minimum, provide a hook which subclasses can use in these cases.

(1) cannot be fixed in uxarray directly because the operation goes directly to xr.Dataset's methods, and when xr.Dataset sees that the second term is an instance of xr.DataArray it decides to handle everything, without giving the second term a chance to alter the behavior in any way.
(2) and (3) cannot be fixed in uxarray directly because numpy's __array_ufunc__ sees that the first term is not a subclass of the second term (or vice versa), so it starts trying to apply __array_ufunc__ in order of inputs. This leads to applying the xarray object's __array_ufunc__, which similarly decides to handle everything after seeing that the second term is an instance of an xarray object (xr.DataArray in (2); xr.Dataset in (3)), without giving the second term a chance to alter the behavior in any way.

What did you expect to happen?

I expected these cases to all return objects with type UxDataset.

Can you provide a MCVE to repoduce the bug?
import numpy as np
import uxarray as ux

uxds = ux.tutorial.open_dataset('quad-hexagon')
uxda = uxds['t2m']
xrds = uxds.to_xarray()
xrda = uxda.to_xarray()

# case 1
print(type(xrds * uxda))  # expected UxDataset, got xr.Dataset
# case 2
print(type(np.minimum(xrds, uxda)))  # expected UxDataset, got xr.Dataset
# case 3
print(type(np.maximum(xrda, uxds)))  # expected UxDataset, got xr.Dataset
Vorherrschende Sprache
Python
Sterne
231
Forks
58
Ø Merge
6 T. 7 Std.
Gemergte PRs (30 T.)
9

Entwicklungsumgebung

Erste Schritte

  1. Lesen Sie das ganze Issue und danach den Beitragsleitfaden des Projekts.
  2. Schreiben Sie ins Issue, dass Sie es übernehmen — das erspart doppelte Arbeit.
  3. Forken Sie das Repository und arbeiten Sie in einem Branch.
  4. Öffnen Sie einen Pull Request, der die Issue-Nummer nennt.

Mehr aus UXARRAY/uxarray

Alle Issues in UXARRAY/uxarray

Ähnliche Issues

Weitere Issues zu Python

Neue Issues direkt in Ihr Postfach

Eine kurze Übersicht über anfängerfreundliche GitHub-Issues.