v0.2.3

What's Changed

• Export gcs_refreshable_credentials in __init__.py by @lindseynield in #776
• Add icechunk.print_debug_info() and Issue Templates by @ianhi in #753
• Make repositories pickelable by @mpiannucci in #778
• Add repr for Storage by @mpiannucci in #782
• Update changelog for 0.2.3 release by @mpiannucci in #783

Full Changelog: v0.2.2...v0.2.3

v0.2.2

What's Changed

• Fix: Check for clean prefix hides access errors by @paraseba in #766
• Only write content-type if the object store accepts metadata by @paraseba in #769
• feat: unify API on snapshot_id by @DahnJ in #770
• readonly_session("main", as_of=...) to open a branch at a timestamp by @paraseba in #768
• doc: fix wrong name in docstrings by @ianhi in #775
• Add support for get_credentials callable in gcs_storage by @lindseynield in #774
• Release prep for 0.2.2 by @mpiannucci in #771

New Contributors

• @DahnJ made their first contribution in #770
• @lindseynield made their first contribution in #774

Full Changelog: v0.2.1...v0.2.2

v0.2.1

What's Changed

• Add optional overrides for nonconforming object stores by @paraseba in #763
• Release version 0.2.1 by @paraseba in #765

Full Changelog: v0.2.0...v0.2.1

v0.2.0

What's Changed

• Docs: fix errors in quickstart by @ianhi in #727
• Add ERA5 demo dataset / Benchmarks on coiled by @dcherian in #684
• Add optional object store backend for s3 by @mpiannucci in #728
• Add Virtual and Configuration docs by @mpiannucci in #720
• Update xarray.md by @NicWayand in #732
• Skip sharding codec in test by @paraseba in #735
• Repository.readonly_session default by @paraseba in #737
• Benchmarks incremental improvements by @dcherian in #740
• Update xarray and zarr docs by @ianhi in #739
• Remote benchmark tweaks by @dcherian in #741
• Flatbuffers for serialization by @paraseba in #733
• Fix getsize benchmark by @dcherian in #742
• Python's Repository.ancestry now returns an iterator by @paraseba in #751
• Treat zarr metadata as a blob (mostly) by @paraseba in #749
• DOC: update contributing instructions by @ianhi in #748
• Switch to conditional PUT for branch updates by @paraseba in #755
• Fix for storage test by @mpiannucci in #760
• Fix bug in clean prefix detection by @paraseba in #761
• New API to set multiple virtual chunk refs at once by @paraseba in #757
• Release version 0.2.0 by @paraseba in #758

New Contributors

• @ianhi made their first contribution in #727
• @NicWayand made their first contribution in #732

Full Changelog: v0.1.3...v0.2.0

v0.1.3

What's Changed

• Change yml dependency by @paraseba in #704
• Don't write dimension_names to JSON if null by @dcherian in #709
• Add GCS Refreshable Credentials by @mpiannucci in #687
• Icechunk learns how to do diff and status by @paraseba in #711
• diff and status return richer information by @paraseba in #715
• Diff should not include changes applied in the from commit by @paraseba in #719
• Add static bearer token auth for GCS by @mpiannucci in #721
• Release 0.1.3 by @mpiannucci in #722

Full Changelog: v0.1.2...v0.1.3

v0.1.2

What's Changed

• Add Storage specific docs by @mpiannucci in #666
• Optionally expire refs when runing snapshot expiration by @paraseba in #673
• Add conda-forge badge & installation instructions for conda by @andersy005 in #674
• Better Storage errors by @paraseba in #680
• Add multithreaded test by @paraseba in #678
• Add example code for tigris by @dcherian in #688
• Fix preload manifest regexes by @paraseba in #692
• Better error reporting and logs by @paraseba in #689
• Add logging to complex operations by @paraseba in #693
• Expose spec version to python by @paraseba in #696
• Faster getsize_prefix by @paraseba in #697
• Faster exists method in the Store by @paraseba in #701
• Add --skip-setup flag to benchmarks runner by @dcherian in #699
• Release version 0.1.2 by @paraseba in #695

New Contributors

• @andersy005 made their first contribution in #674

Full Changelog: v0.1.1...v0.1.2

icechunk-v0.1.1

What's Changed

• Update sample datasets documentation by @paraseba in #662
• [Docs] remove GitHub pages build by @jhamman in #661
• Fix credential fetcher deserialization by @paraseba in #671
• Release version 0.1.1 by @paraseba in #672

Full Changelog: icechunk-v0.1.0...v0.1.1

icechunk-v0.1.0

What's Changed

• Tag delete design document by @paraseba in #621
• Experimental support for Tigris by @paraseba in #625
• Icechunk learns how to delete tags. by @paraseba in #626
• Fix storage doc string by @abarciauskas-bgse in #628
• doc: build docs on readthedocs by @jhamman in #254
• Uncache history by @paraseba in #635
• Expiration algorithm design document by @paraseba in #631
• Implement expiration of snapshots by @paraseba in #636
• Update some notebooks to latest syntax by @mpiannucci in #639
• Fix bug in manifest splitting by @dcherian in #642
• Python expiration and GC by @paraseba in #641
• Snapshot timestamp checks by @paraseba in #643
• Update rand dependency by @paraseba in #644
• Implement Python's Repository.async_ancestry by @paraseba in #645
• Remove some cruft from the on disk datastuctures by @paraseba in #647
• Update store tests for zarr 3.0.2 by @mpiannucci in #655
• Add First Pass at Version Control Docs by @mpiannucci in #650
• Fetch initialized chunks by @paraseba in #651
• Require opt-in to pickling writable Sessions and Stores. by @dcherian in #652
• Fix pickling tests by @dcherian in #657
• Enable Rust-ci workflow on python changes by @paraseba in #658
• Implement manifest preloading by @paraseba in #649
• Fix benchmarks for zarr 3.0.2 by @dcherian in #656
• Don't create a repository unless its prefix is empty by @paraseba in #659
• Manifest split design document by @paraseba in #593
• Upgrade xarray.to_icechunk and related docs by @dcherian in #633
• Release version 0.1.0 by @paraseba in #660

Full Changelog: icechunk-v0.1.0-alpha.15...icechunk-v0.1.0

icechunk-v0.1.0-alpha.15

What's Changed

• Better binary format constants by @paraseba in #588
• Fix link to Arraylake version control docs by @hendrikmakait in #587
• Mechanism to deserialize different spec versions by @paraseba in #589
• group.array -> create_array by @paraseba in #597
• Get repository config by @paraseba in #594
• Bump serde_json from 1.0.135 to 1.0.137 in the rust-dependencies group by @dependabot in #596
• Evict asset caches based on size by @paraseba in #595
• Fix codespell by @dcherian in #607
• Fix error on: from icechunk import * by @paraseba in #603
• Benchmark with pytest-benchmark by @dcherian in #552
• Efficient getsize in the Zarr store by @paraseba in #605
• Sync docs by @mpiannucci in #612
• Better defaulting mechanism for configuration by @paraseba in #613
• Generate one manifest per array by @paraseba in #598
• Release 0.1.0-alpha.13 by @paraseba in #614
• Release 0.1.0-alpha.14 by @paraseba in #615
• Release 0.1.0-alpha.15 by @paraseba in #617

New Contributors

• @hendrikmakait made their first contribution in #587

Full Changelog: icechunk-v0.1.0-alpha.12...icechunk-v0.1.0-alpha.15

icechunk-v0.1.0-alpha.12

What's Changed

• Add tests to check no double asset download by @paraseba in #568
• Wait for azurite container before workflow by @paraseba in #569
• Disallow deleting main branch by @paraseba in #571
• Add validate_container argument to set_virtual_ref by @paraseba in #573
• Better ancestry interface by @paraseba in #574
• Mutable python config by @paraseba in #577
• Add helper Repository.reopen by @paraseba in #578
• More time for docker to initialize by @paraseba in #580
• Split and download virtual chunks using concurrent requests by @paraseba in #579
• Concurrently download bytes by @paraseba in #582
• Some initial work towards better python repr by @paraseba in #584
• Release 0.1.0-alpha-12 by @paraseba in #586

Full Changelog: icechunk-v0.1.0-alpha.11...icechunk-v0.1.0-alpha.12