Feature/accessor improvement

[FBumann]

• Closes Improve IDE/Editor Support for External Accessors #11078
• Tests added
• User visible changes (including notable bug fixes) are documented in whats-new.rst
• New functions/methods are listed in api.rst

Summary

Add typed properties for external accessor packages (hvplot, cf-xarray, pint-xarray, rioxarray, xarray-plotly), enabling full IDE support including autocompletion, parameter hints, docstrings, and go-to-definition.

Problem

External accessor packages currently have no IDE support:

import xarray as xr
import hvplot.xarray  # required just to register accessor

ds = xr.open_dataset("data.nc")

ds.plot.scatter()    # Built-in: full IDE completion ✓
ds.hvplot.scatter()  # External: nothing. no hints, no docs ✗

Feature	Built-in .plot	External Accessor
Method completion	✓	✗
Parameter hints	✓	✗
Docstrings	✓	✗
Go-to-definition	✓	✗

This affects every accessor package: hvplot, cf-xarray, pint-xarray, rioxarray, xarray-plotly, and the entire xarray-contrib ecosystem.

Solution

Add typed properties with lazy imports on DataArray, Dataset, and DataTree:

# In xarray/core/dataarray.py
if TYPE_CHECKING:
    from hvplot.xarray import hvPlotAccessor

class DataArray:
    @property
    def hvplot(self) -> hvPlotAccessor:
        """hvPlot accessor for interactive plotting."""
        from xarray.accessors import DATAARRAY_ACCESSORS, _get_external_accessor
        return _get_external_accessor("hvplot", self, DATAARRAY_ACCESSORS)

Key behaviors

1. hasattr() returns False for uninstalled packages - Clean runtime introspection

   hasattr(da, 'hvplot')  # False (if hvplot not installed)
   hasattr(da, 'plotly')  # True (if xarray-plotly installed)

2. Helpful error messages when accessing uninstalled accessors:

   da.hvplot
   # AttributeError: 'DataArray' object has no attribute 'hvplot'.
   # Install with: pip install hvplot

3. External packages don't overwrite the typed properties - xarray's register_*_accessor skips registration for known external accessors, preserving IDE support.

4. Full IDE support for installed packages - autocompletion, parameter hints, docstrings all work.

Supported accessors

Accessor	Package	DataArray	Dataset	DataTree
hvplot	hvplot	✓	✓	✓
cf	cf-xarray	✓	✓	✓
pint	pint-xarray	✓	✓
rio	rioxarray	✓	✓
plotly	xarray-plotly	✓	✓

Files changed

• xarray/accessors.py (new) - Accessor registry and lazy loader
• xarray/core/dataarray.py - Added typed accessor properties
• xarray/core/dataset.py - Added typed accessor properties
• xarray/core/datatree.py - Added typed accessor properties
• xarray/core/extensions.py - Skip registration for known external accessors
• pyproject.toml - Added mypy ignores for external packages

Backward compatibility

• Fully backward compatible - Existing code continues to work unchanged
• External packages using register_*_accessor continue to work (registration is silently skipped for known accessors)
• Users who don't use external accessors see no difference

Example usage

import xarray as xr

da = xr.DataArray([1, 2, 3], dims=['x'])

# IDE now provides full autocompletion for installed packages!
da.hvplot.line()  # If hvplot installed: works with full IDE support
da.cf.describe()  # If cf-xarray installed: works with full IDE support

# Helpful errors for uninstalled packages
da.rio  # AttributeError with install instructions

Related

• Follows the pattern established by dask.distributed
• Addresses IDE support gap identified in PR Feature/plotly accessor #11075

[welcome]

Thank you for opening this pull request! It may take us a few days to respond here, so thank you for being patient.
If you have questions, some answers may be found in our contributing guidelines.