[XXXXXQ-0206]

Summary

arr[[]] raises IndexError: unsupported selection item for basic indexing. The same selection works as an empty integer array, and NumPy accepts it:

a = zarr.create_array({}, data=np.arange(16).reshape(4, 4), chunks=(2, 2))
a[[]]                            # IndexError
a[np.array([], dtype=np.intp)]   # (0, 4)
np.arange(16).reshape(4, 4)[[]]  # (0, 4)

IntSequence (declared at src/zarr/core/indexing.py:42) includes list[int], so an empty list is a valid selection. The empty array form already returns (0, 4); only the list form is rejected.

Cause: is_integer_list (:233) and is_bool_list (:238) both require len(x) > 0, so an empty list falls through to BasicIndexer, which accepts only int and slice.

The change drops len(x) > 0 from is_integer_list only, and makes replace_lists use np.intp for an empty list (it otherwise infers float64). is_bool_list keeps its guard, so an empty list stays an empty integer selection rather than an empty mask, as in NumPy.

The new test covers empty lists in bare, tuple, oindex and vindex forms, plus the empty-array, boolean-list and write controls. It fails on the base revision (src blob 8ea59eb9 -> 2926551a); tests/test_indexing.py goes from 592 to 594 passed. ruff check / ruff format --check are clean.

If you would rather keep an empty list raising, we are happy to adjust.

For reviewers

The boolean-list guard is deliberately unchanged — please check that leaving is_bool_list untouched is how you want [] disambiguated.

Author attestation

• I am a human, these are my changes, and I have reviewed and understood every change and can explain why each is correct.

TODO

• Add unit tests and/or doctests in docstrings
• Add docstrings and API docs for any new/modified user-facing classes and functions
• New/modified features documented in docs/user-guide/*.md
• Changes documented as a new file in changes/
• GitHub Actions have all passed
• Test coverage is 100% (Codecov passes)

[Copilot]

Copilot was unable to review this pull request because the user who requested the review has reached their quota limit.

An organization owner can choose the repository organization as the billing source for reviews requested by members with a paid Copilot license. Learn about Copilot code review billing.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 94.71%. Comparing base (ec2c359) to head (8b3d44e).

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #4518   +/-   ##
=======================================
  Coverage   94.71%   94.71%           
=======================================
  Files          94       94           
  Lines       13663    13663           
=======================================
  Hits        12941    12941           
  Misses        722      722           

Files with missing lines	Coverage Δ
src/zarr/core/indexing.py	96.57% <100.00%> (ø)

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[d-v-b]

don't nest function definitions without a reason

[d-v-b]

consider this instead:

def replace_lists(selection: SelectionNormalized) -> SelectionNormalized:
    # An empty list has no dtype to infer; NumPy indexes with it as an empty integer array.
    return tuple(
        cast("ArrayOfIntOrBool", np.asarray(dim_sel, dtype=np.intp if len(dim_sel) == 0 else None))
        if isinstance(dim_sel, list)
        else dim_sel
        for dim_sel in selection
    )

[XXXXXQ-0206]

Addressed both review points in 8b3d44e: the empty-list cast is inlined in replace_lists (no nested function), using the dtype rule you suggested (np.intp for an empty dimension). tests/test_indexing.py: 594 passed, 1 skipped, 2 xfailed; ruff check and ruff format --check are clean. Could you take another look when you have a moment?