v0.5.0

What's Changed

• new release note section by @keewis in #264
• colorize the pytest output by @keewis in #265
• drop python 3.9 by @keewis in #266
• implement the PintIndex by @keewis in #163
• Pass kwargs and assume_sorted to xarray interp for both DataArray and Dataset by @martijnvandermarel in #270
• hook maintenance by @keewis in #296
• ignore bots in the auto-generated release notes by @keewis in #297
• release 0.5 by @keewis in #301

New Contributors

• @martijnvandermarel made their first contribution in #270

Full Changelog: v0.4...v0.5.0

v0.5a0

Most notably this includes the implementation of a PintIndex, which technically is a breaking change (hence the alpha release first).

What's Changed

• new release note section by @keewis in #264
• colorize the pytest output by @keewis in #265
• drop python 3.9 by @keewis in #266
• implement the PintIndex by @keewis in #163
• Pass kwargs and assume_sorted to xarray interp for both DataArray and Dataset by @martijnvandermarel in #270
• hook maintenance by @keewis in #296
• ignore bots in the auto-generated release notes by @keewis in #297

New Contributors

• @martijnvandermarel made their first contribution in #270

Full Changelog: v0.4...v0.5a0

v0.4

Bug fixes

• ignore datetime units on attributes (#241)

Maintenance

• test python 3.11 and 3.12 in CI (#228, #262)
• adopt SPEC0 (#228): bump dependency versions:
  • python to 3.9
  • xarray to 2022.06.0
  • numpy to 1.23
  • pint to 0.21

v0.3

New features

• allow using .pint.quantify as a identity operator (#47)

Bug fixes

• preserve MultiIndex objects in .pint.dequantify (#164)
• allow quantifying dimension coordinates (#105)

Maintenance

• drop support for python 3.7 (#153)
• test python 3.10 in CI (#155)

v0.2.1

This release brings a few docs improvements and a way to tell .pint.quantify not to quantify a certain variable.

v0.2

New features:

• provide more context about variables when reporting unit errors
• pseudo-support units in indexes
• unit-aware versions of sel, loc, drop_sel, chunk, reindex, reindex_like, interp, interp_like, ffill, bfill, and interpolate_na
• allow converting all data variables to the same unit using Dataset.pint.to
• a small tutorial show-casing the basic functionality

Bug fixes:

• parse units using parse_units instead of parse_expression

Release v0.1

The initial release of pint-xarray.