5 Pull requests merged by 3 people

• Ignore stale children when reconsolidating metadata

  #2980 merged May 31, 2025

• Use automatic chunking in array creation routines

  #3103 merged May 30, 2025

• Add a GroupNotFoundError

  #3066 merged May 29, 2025

• remove insertion of vlen-string codec for v2 metadata creation

  #3100 merged May 28, 2025

• Test using latest numpy

  #3064 merged May 28, 2025

7 Pull requests opened by 4 people

• Fix release note links

  #3106 opened May 31, 2025

• Bump redis from 6.1.0 to 6.2.0 in the requirements group

  #3108 opened Jun 2, 2025

• chore: update pre-commit hooks

  #3109 opened Jun 2, 2025

• Use context manager for temporary directories in store tests

  #3110 opened Jun 3, 2025

• Cleanups to implementation of `create()`

  #3111 opened Jun 3, 2025

• Fix order handling

  #3112 opened Jun 3, 2025

• Add pytest pin

  #3113 opened Jun 3, 2025

3 Issues closed by 2 people

• zarr.consolidated_metadata should not use previously consolidated (stale) metadata

  #2921 closed May 31, 2025

• Mapping of new dtypes to `numpy` dtypes

  #3101 closed May 29, 2025

• Raise specific errors when failing to open a group.

  #2298 closed May 29, 2025

5 Issues opened by 4 people

• [Unexpected Behaviour] `open_group` raises `AssertionError` if array w/o group exists

  #3114 opened Jun 3, 2025

• Monthly issue metrics report

  #3107 opened Jun 1, 2025

• Add a store.read_only setter method

  #3105 opened May 30, 2025

• consider removing default compressor / filters / serializer from config

  #3104 opened May 30, 2025

• zarr python 2.18 is incompatible with numpy's variable-length strings dtype

  #3102 opened May 28, 2025

21 Unresolved conversations

Sometimes conversations happen on old items that aren’t yet closed. Here is a list of all the Issues and Pull Requests with unresolved conversations.

• Add Pyodide support and CI jobs for Zarr

  #1903 commented on Jun 3, 2025 • 19 new comments

• Support async FSMap objects in zarr.open

  #2774 commented on Jun 3, 2025 • 11 new comments

• refactor v3 data types

  #2874 commented on Jun 3, 2025 • 9 new comments

• Coalesce and parallelize partial shard reads

  #3004 commented on Jun 3, 2025 • 4 new comments

• Make `create_array` signatures consistent

  #2819 commented on May 31, 2025 • 1 new comment

• Regression testing

  #3099 commented on May 29, 2025 • 0 new comments

• refactor warnings

  #3098 commented on May 31, 2025 • 0 new comments

• Fix typing in store tests

  #3097 commented on May 29, 2025 • 0 new comments

• Add async oindex and vindex methods to AsyncArray

  #3083 commented on May 29, 2025 • 0 new comments

• should we abstract over v2 and v3 codecs

  #2654 commented on Jun 2, 2025 • 0 new comments

• Support Blosc2 codec

  #1896 commented on Jun 2, 2025 • 0 new comments

• "Release notes" link on PyPI is 404

  #3073 commented on May 31, 2025 • 0 new comments

• [v3] Structured dtype support

  #2134 commented on May 31, 2025 • 0 new comments

• how to abstract over v2 / v3 metadata differences

  #3080 commented on May 31, 2025 • 0 new comments

• deprecate / remove support for creating v2 array metadata with `null` fill value

  #3084 commented on May 31, 2025 • 0 new comments

• Device support in `zarr-python` (especially for GPU)

  #2658 commented on May 31, 2025 • 0 new comments

• improve control over warnings

  #3096 commented on May 31, 2025 • 0 new comments

• make deprecated positional arguments keyword-only

  #3092 commented on May 31, 2025 • 0 new comments

• can we reject dtype inference for numpy object arrays

  #3077 commented on May 31, 2025 • 0 new comments

• Refactor CodecPipeline for flexibility

  #3051 commented on May 31, 2025 • 0 new comments

• accelerating v2 -> v3 migration

  #3076 commented on May 30, 2025 • 0 new comments