v0.4.1

v0.4.1 (2024-11-07)

Contributors to this version: Gabriel Rondeau-Genesse (@RondeauG), Trevor James Smith (@Zeitsperre), Louise Arnal (@lou-a).

New features and enhancements

• A periods parameter was added to frequency_analysis.local.fit to compute multiple separate periods with a single call. (PR/219).
• In xhydro.cc.sampled_indicators, the delta_type argument can now be a dictionary or None, in which case the attribute delta_kind is used. (PR/220).
• In xhydro.cc.sampled_indicators, weights along a time or horizon dimension will no longer reduce that dimension. (PR/220).

Bug fixes

• Fixed a bug in xhydro.modelling.format_input where the function would fail if the input data was a dask array. (PR/214).
• The executable parameter in the Hydrotel model class is now always required. (PR/214).

Breaking changes

• The xhydro testing utilities have been rewritten to use pooch for downloading and caching testing datasets from hydrologie/xhydro-testdata. (PR/212).
• The xhydro testing utilities now require pytest-xdist as a development dependency. (PR/212).
• Many core dependencies have been updated to more modern versions. (PR/218, PR/228).
• The delta_type argument in xhydro.cc.sampled_indicators has been renamed to delta_kind and is no longer positional. (PR/220).
• The xhydro.pmp module has been moved to xhydro.indicators.pmp. (PR/226).

Internal changes

• Updated the notebooks to reduce the load on ReadTheDocs. (PR/211).
• Tests can now be run using the pytest-xdist plugin for distributed testing. See the pytest-xdist documentation <https://pytest-xdist.readthedocs.io/en/stable/>_ for more information. (PR/212).
• Several tests reliant on online servers and services have been marked as online to prevent them from running in contexts where internet access is limited. (PR/212).
• Many function docstrings and type hints have been updated for accuracy and precision. (PR/212).
• The xHydro project now has a set of logos for use in documentation, the readme, and other materials. (PR/217).
• xhydro.indicators is now a package with submodules for different types of indicators. Previous functions have been moved to a generic submodule and are imported at the top level for backwards compatibility. (PR/226).
• xh.testing.utils.publish_release_notes now has a latest parameter to print only the latest release notes. (PR/228).

v0.4.0

Contributors to this version: Trevor James Smith (@Zeitsperre), Gabriel Rondeau-Genesse (@RondeauG), Thomas-Charles Fortier Filion (@TC-FF).

New features and enhancements

• xhydro now supports RavenPy v0.15.0 (RavenHydroFramework v3.8.1). (PR/161).
• Regional frequency analysis functions as well as Resampling function for uncertainties have been added to the xhydro.frequency_analysis module. (PR/186).
• New function xhydro.modelling.format_input to format CF-compliant input data for hydrological models (currently only supports Hydrotel). (PR/185).

Internal changes

• numpy has been pinned below v2.0.0 until xclim and other dependencies are updated to support it. (PR/161).
• A helper script has been added in the CI directory to facilitate the translation of the xhydro documentation. (GH/63, PR/163).
• The conda environment now relies on the newly created xdatasets package. (PR/164).
• The cookiecutter has been updated to the latest commit. Changes include workflow fixes, stricter coding standards, and many small adjustments to the documentation. (PR/164).
• A previously uncaught YAML formatting issue has been addressed. Stricter style conventions are now enforced. (PR/174).
• Chunking was adjusted in a few functions to work with the new requirements of apply_ufunc. (PR/180).
• Updated the cookiecutter template to the latest commit. (PR/177):
  • Actions have been updated and synchronized.
  • Warnings in Pull Requests from forks are now less buggy.
  • A new pre-commit hook and linting step for validating numpy docstrings has been added (numpydoc).
  • All pip-based dependencies used to run in CI are now managed by a CI/requirements_ci.txt that uses hashes of packages for security.
• Added two new Batch (.bat) files to help facilitate the translation of and the generation of the xhydro documentation in Windows environments. (PR/196).
• The bumpversion workflow now uses the Hydrologie Helper Bot to make signed commits. (PR/199).
• Updated the cookiecutter template to the latest commit. (PR/199):
  • Updated development dependencies to the latest versions.
  • Staged support for Python3.13.
  • Added environment caching to existing workflows.

Breaking changes

• xhydro now requires python >= 3.10. (PR/195).

v0.3.6

v0.3.6 (2024-06-10)

Contributors to this version: Gabriel Rondeau-Genesse (@RondeauG), Richard Arsenault (@richardarsenault), Sébastien Langlois (@sebastienlanglois).

New features and enhancements

• Added support for the Hydrotel hydrological model. (PR/18).
• Added support for various hydrological models emulated through the Raven hydrological framework. (PR/128).
• Added optimal interpolation functions for time-series and streamflow indicators. (PR/88, PR/129).
• Added optimal interpolation notebooks. (PR/123).
• Added surface properties (elevation, slope, aspect ratio) to the gis module. (PR/151).

Breaking changes

• Hydrological models are now classes instead of functions and dictionaries. (GH/93, PR/18).
• xhydro now uses a 'src' layout <https://packaging.python.org/en/latest/discussions/src-layout-vs-flat-layout>_ for the package. (PR/147).

Internal changes

• Tests using the gamma distribution were changed to the gumbel_r to avoid changes in xclim v0.49.0. (PR/145).
• The cookiecutter template has been updated to the latest commit. Changes include the addition of a CODE_OF_CONDUCT.rst file, the renaming of CHANGES.rst to CHANGELOG.rst, and many small adjustments to the documentation. (PR/147).
• Added a CODE_OF_CONDUCT.rst file with Contributor Covenant guidelines. (PR/147).

v0.3.5

v0.3.5 (2024-03-20)

Contributors to this version: Trevor James Smith (@Zeitsperre), Thomas-Charles Fortier Filion (@TC-FF), Sébastien Langlois (@sebastienlanglois), Gabriel Rondeau-Genesse (@RondeauG).

New features and enhancements

• xhydro has implemented a gis module that facilitates geospatial tasks needed for gathering hydrological inputs. (GH/60, PR/61).

Internal changes

• Added a workflow based on actions/labeler to automatically label Pull Requests based on files changed. (PR/68).
• Added a conditional trigger to the test-notebooks job to run in advance of pull request approval in the event that the notebooks found within docs/notebooks have been modified (labeled "notebooks"). (PR/68).
• Significant changes to the Continuous Integration (CI) setup. (PR/65):
  • Added a workflow configuration using label_on_approval.yml and modifications of main.yml so that fewer tests are run on Pull Requests before they are fully approved.
  • Added some pre-commit configurations to both clean up the code within notebooks (NbQA) and strip their outputs (nbstripout).
  • tox is now fully v4.0-compliant.
  • Added a Makefile recipe to facilitate installation of esmpy when esmf is installed and visible on the $PATH.
  • Added a Makefile recipe for running tests over Jupyter notebooks.
  • Synchronized dependencies between pyproject.toml and conda configuration files.
• Moved the notebooks under a Usage section in the documentation. (GH/114, PR/118).

v0.3.4

v0.3.4 (2024-02-29)

Contributors to this version: Trevor James Smith (@Zeitsperre), Thomas-Charles Fortier Filion (@TC-FF), Gabriel Rondeau-Genesse (@RondeauG).

New features and enhancements

• Added French language support to the documentation. (GH/53, PR/55).
• Added a new set of functions to support creating and updating pooch registries, caching testing datasets from hydrologie/xhydro-testdata, and ensuring that testing datasets can be loaded into temporary directories. (PR/62).
• xhydro is now configured to use pooch to download and cache testing datasets from hydrologie/xhydro-testdata. (PR/62).
• xhydro is now Semantic Versioning v2.0.0 <https://semver.org/spec/v2.0.0.html>_ compliant. (PR/70).
• Added new functions to xhydro.frequency_analysis.local to calculate plotting positions and to prepare plots. (PR/87).
• xscen now supports Python3.12. (PR/99).
• xscen now supports pandas >= 2.2.0, xarray >= 2023.11.0, and xclim >= 0.47.0. (PR/99).
• Added xh.cc.sampled_indicators to compute future indicators using a perturbation approach and random sampling. (PR/54).

Breaking changes

• Added pooch as an installation dependency. (PR/62).
• xhydro now requires xarray>=2023.11.0, xclim>=0.48.2, xscen>=0.8.3, and, indirectly, pandas>=2.2.0. The main breaking change is in how yearly frequencies are called ('YS-' instead of 'AS-'). (PR/54).
• Functions that output a dict with keys as xrfreq (namely, xh.indicators.compute_indicators) will now return the new nomenclature (e.g. "YS-JAN" instead of "AS-JAN"). (PR/54).

Internal changes

• Added a new module for testing purposes: xhydro.testing.helpers with some new functions. (PR/62):
  • generate_registry: Parses data found in package (xhydro.testing.data), and adds it to the registry.txt
  • load_registry: Loads installed (or custom) registry and returns dictionary
  • populate_testing_data: Fetches the registry and optionally caches files at a different location (helpful for pytest-xdist).
• Added a pre-commit hook (numpydoc) to ensure that numpy docstrings are formatted correctly. (PR/62).
• The cookiecutter has been updated to the latest commit (PR/70, PR/106):
  • Added some workflows (Change file labelling, Cache cleaning, Dependency scans, OpenSSF Scorecard).
  • The README has been updated to organize badges in a table, including a badge for the OpenSSF Scorecard.
  • Updated pre-commit hook versions to the latest available.
  • Formatting tools are now pinned to their pre-commit equivalents.
  • actions-version-updater.yml has been replaced by dependabot.
  • Addressed a handful of misconfigurations in the workflows.
  • Updated ruff to v0.2.0 and black to v24.2.0.
• Added a few functions missing from the API to their respective modules via __all__. (PR/99).

Release v0.3.0

v0.3.0 (2023-12-01)

Contributors to this version: Gabriel Rondeau-Genesse (@RondeauG), Trevor James Smith (@Zeitsperre).

New features and enhancements

• The xhydro planification was added to the documentation. (GH/39, PR/49).

Breaking changes

• xhydro now adheres to PEPs 517/518/621 using the flit backend for building and packaging. (PR/50).

Bug fixes

• The return_level dimension in xh.frequency_analysis.local.parametric_quantiles() is now the actual return level, not the quantile. (GH/41, PR/43).

Internal changes

• Added xhydro.testing.utils.publish_release_notes() to help with the release process. (PR/37).
• xh.frequency_analysis.local.parametric_quantiles() and xh.frequency_analysis.local.criteria() are now lazier. (GH/41, PR/43).
• The cookiecutter template has been updated to the latest commit via cruft. (PR/50):
  • Manifest.in and setup.py have been removed.
  • pyproject.toml has been added, with most package configurations migrated into it.
  • HISTORY.rst has been renamed to CHANGES.rst.
  • actions-version-updater.yml has been added to automate the versioning of the package.
  • bump-version.yml has been added to automate patch versioning of the package.
  • pre-commit hooks have been updated to the latest versions; check-toml and toml-sort have been added to cleanup the pyproject.toml file.
  • ruff has been added to the linting tools to replace most flake8 and pydocstyle verifications.

v0.2.0

v0.2.0 (2023-10-10)

Contributors to this version: Trevor James Smith (@Zeitsperre), Gabriel Rondeau-Genesse (@RondeauG), Thomas-Charles Fortier Filion (@TC-FF), Sébastien Langlois (@sebastienlanglois)

Announcements

• Support for Python3.8 and lower has been dropped. (PR/11).
• xHydro now hosts its documentation on Read the Docs. (GH/22, PR/26).
• Local frequency analysis functions have been added under a new module xhydro.frequency_analysis. (PR/20, PR/27).

New features and enhancements

• GitHub Workflows for automated testing using tox have been added. (PR/11).
• Support for various xscen functions has been added to compute indicators and various climate change metrics. (PR/21).
• New function xh.indicators.compute_volume to convert streamflow data to volumes. (PR/20, PR/27).
• New function xh.indicators.get_yearly_op to compute block operation (e.g. block maxima, minima, etc.). (PR/20, PR/27).

Breaking changes

• xHydro repository has renamed its primary development branch from master to main. (PR/13).
• xHydro now requires a conda environment to be installed. (PR/21).

Bug fixes

• N/A

Internal changes

• Added a Pull Request template. (PR/14).
• Various updates to the autogenerated boilerplate (Ouranosinc/cookiecutter-pypackage) via cruft. (PR/11, PR/12, PR/13):
  • General updates to pre-commit hooks, development dependencies, documentation.
  • Added configurations for Pull Request and Issues templates, Zenodo.
  • Documentation now makes use of sphinx directives for usernames, issues, and pull request hyperlinks (via sphinx.ext.extlinks). (GH/15).
  • GitHub Workflows have been added for automated testing, and publishing.
  • Some sphinx extensions have been added/enabled (sphinx-codeautolink, sphinx-copybutton).
  • Automated testing with tox now updated to use v4.0+ conventions.
  • Removed all references to travis.ci.
• Deployments to TestPyPI and PyPI are now run using GitHub Workflow Environments as a safeguarding mechanism. (PR/28).
• Various cleanups of the environment files. (GH/23, PR/30).
• xhydro now uses the trusted publishing mechanism for PyPI and TestPyPI deployment. (PR/32).
• Added tests. (PR/27).

v0.1.2

What's Changed

• Generate docs by @TC-FF in #5
• Prepare v0.1.2 by @TC-FF in #10

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