v0.12.0

Overview

This release introduces BI500 parsing support, Match Geometry regridding, improvements to existing processing functions, and a range of bug fixes. It also modernizes Python packaging and dependency support through Scientific Python SPEC 0, adds Python 3.14 support, and strengthens the CI and test infrastructure.

Thanks to @LOCEANlloydizard, @praneethratna, @leewujung, @beatfactor, @ctuguinay, @gavinmacaulay, @dash-uvic, @binggao1230, and @eliascapriles-NOAA for their contributions!

New functionalities

• Add parser support for BI500 files by @praneethratna in #1252
• Add regridding based on Echoview Match Geometry by @eliascapriles-NOAA in #1597
• Support parsing and storing selected channels by @praneethratna in #1756
• Add preliminary AZFP ULS6 support and ULS5 improvements by @dash-uvic in #1581

Enhancements and performance

• Improve performance of noise reduction and seafloor detection by @LOCEANlloydizard in #1780
• Refactor logging, introduce package-level verbosity control, and clean up the logging infrastructure by @praneethratna in #1740 and #1772
• Group EK60 and EK80 sonar model families by @praneethratna in #1738
• Prepare the deprecation of combine_echodata by @ctuguinay in #1725

Bug and warning fixes

• Fix SNR threshold typing and xarray scalar extraction by @beatfactor in #1622
• Fix Blackwell depth-grid selection by @LOCEANlloydizard in #1661
• Fix file-permission checks for target stores by @binggao1230 in #1676
• Handle NaT values in time encoding by @LOCEANlloydizard in #1694
• Handle duplicate timestamps in add_location by @beatfactor in #1627
• Fix channel-frequency mapping in compute_MVBS by @LOCEANlloydizard in #1698
• Pad vendor parameters with unequal lengths across channels by @leewujung in #1702
• Check the first datagram type when identifying EK60/EK80 models by @leewujung in #1724
• Fix AZFP ULS5 single-frequency file handling by @dash-uvic in #1714
• Prevent crashes on zero-length dimensions during EchoData conversion by @beatfactor in #1624
• Fix ECS calibration channel ordering when broadband calibration data is absent by @ctuguinay in #1768
• Fix BI500 channel identifiers by @LOCEANlloydizard in #1793
• Fix xarray coarsen reductions by @LOCEANlloydizard in #1796
• Update Zarr v3 compressor handling by @leewujung in #1657
• Update string encoding to object type by @leewujung in #1658
• Expose the qc module at the package level by @leewujung in #1771

Dependencies and packaging

• Migrate package configuration to pyproject.toml by @LOCEANlloydizard in #1667
• Adopt Scientific Python SPEC 0 to maintain supported minimum Python and dependency versions by @LOCEANlloydizard in #1739
• Add Python 3.14 support by @LOCEANlloydizard in #1708
• Improve uv development workflow support by @gavinmacaulay in #1705
• Default to Python 3.12 for uv development environments by @gavinmacaulay in #1735
• Update build workflow installation of pyproject.toml extras by @LOCEANlloydizard in #1670

CI, tests, and test data

• Fix Pooch download/cache handling in development setup by @LOCEANlloydizard in #1656
• Split Windows CI into unit and integration jobs by @LOCEANlloydizard in #1680
• Clean up CI workflows by @LOCEANlloydizard in #1723
• Validate data inventory on pushes to main by @LOCEANlloydizard in #1748
• Add initial data-inventory tests by @LOCEANlloydizard in #1684
• Replace MinIO with SeaweedFS by @leewujung in #1798
• Fix MinIO download URL for Windows CI by @LOCEANlloydizard in #1784
• Adapt memory monitoring to new Windows CI jobs by @LOCEANlloydizard in #1737
• Adjust push triggers by @LOCEANlloydizard in #1749
• Restore the PR-test workflow name by @LOCEANlloydizard in #1757
• Ignore Windows service artifacts and clarify test setup by @LOCEANlloydizard in #1750
• Keep local test artifacts outside the repository by @LOCEANlloydizard in #1758
• Remove duplicate EK80 multiplex fixture by @LOCEANlloydizard in #1728
• Add Echoview Match Geometry reference data by @LOCEANlloydizard in #1679
• Add TS-spectrum example data by @LOCEANlloydizard in #1683
• Update TS-spectrum reference-data checksum by @LOCEANlloydizard in #1689
• Improve assertions for interpolated ECS calibration parameters by @LOCEANlloydizard in #1687
• Add heading test data and fixtures by @LOCEANlloydizard in #1704
• Update shared test fixtures by @LOCEANlloydizard in #1707
• Resolve pytest warnings by @gavinmacaulay in #1731
• Add AZFP single-frequency test data by @LOCEANlloydizard in #1733
• Register BI500 test data by @LOCEANlloydizard in #1744
• Streamline PyPI publishing, improve package validation, and modernize conda-forge release checks by @LOCEANlloydizard and @leewujung in #1802

Documentation

• Update links following the organization rename by @LOCEANlloydizard in #1668
• Update contributor documentation for pyproject.toml by @LOCEANlloydizard in #1672
• Fix docstring sections missing from rendered documentation by @gavinmacaulay in #1709
• Restore bibliography in generated documentation by @gavinmacaulay in #1710
• Fix contributing-page reference by @praneethratna in #1711
• Document coerce_increasing_time usage by @leewujung in #1726
• Resolve Sphinx deprecation warnings by @gavinmacaulay in #1736
• Remove outdated GSoC 2024 instructions by @LOCEANlloydizard in #1752
• Rework README badges by @LOCEANlloydizard in #1754
• Resolve documentation build warnings by @gavinmacaulay in #1759
• Fix remaining documentation build warnings and errors by @gavinmacaulay in #1763
• Refine API documentation by @gavinmacaulay in #1764
• Test unpinning Jupyter Book by @LOCEANlloydizard in #1673
• Ignore generated jupyter_execute files by @gavinmacaulay in #1755

Maintenance and automated updates

• Apply automated Scientific Python SPEC 0 minimum-version updates (#1742, #1788, #1795).
• Update GitHub Actions, publishing dependencies, and pre-commit hooks (see full changelog).

New contributors

• @binggao1230 made their first contribution in #1676.
• @eliascapriles-NOAA made their first contribution in #1597.

Full Changelog: v0.11.1...v0.12.0