0.6.0

Full release notes:
https://earthkit-plots.readthedocs.io/en/latest/release-notes/version-0.6-updates.html

0.5.2

What's Changed

• Fixes issue with automatic plotting of polytope data as point clouds by @JamesVarndell in #139

Full Changelog: 0.5.1...0.5.2

0.5.1

Fixed a bug where the lead_time formatter was not working correctly with multiple time values (#134).

0.5.0

What's Changed

Full release notes: https://earthkit-plots.readthedocs.io/en/latest/release-notes/version-0.5-updates.html

• Adds initial implementation of visualising unstructured grids by @JamesVarndell in #14
• Raise useful error when scipy is not installed by @JamesVarndell in #15
• Merge develop into main by @JamesVarndell in #27
• Merge develop into main @ 0.2.7 by @JamesVarndell in #30
• Develop -> main by @JamesVarndell in #78
• Merge develop into main by @JamesVarndell in #127
• Merge develop into main for release 0.5.0 by @JamesVarndell in #133

Full Changelog: 0.3.5...0.5.0

0.4.1

Made the default behaviour of quickplot to not combine vector components by default. This is to avoid unforeseen issues with some data sources. This will be addressed in release 0.6.

0.4.0

What's Changed

• Docs: Add example notebook for interactive plots by @nightowl2626 in #89
• Feature/heatmap by @nightowl2626 in #90
• Feature/windrose by @nightowl2626 in #91
• feat: Add unit aliases by @HCookie in #102
• feat: Add cities method to Figure by @HCookie in #104
• chore: update docs by @HCookie in #111
• feat: Add AND rule for style identities by @HCookie in #108
• Feature/streamplots by @JamesVarndell in #112
• Bugfix/healpy optional dependency by @JamesVarndell in #114
• Fixes a bug which messed up contour lines when suptitle was called by @JamesVarndell in #113
• feat: Vector identification in quickplot by @HCookie in #117
• Image comparison tests by @JamesVarndell in #107
• Bugfix/dont meshgrid xy for 1d data by @EddyCMWF in #97
• Remove auto tricontour over-rides by @EddyCMWF in #98
• fix: Barbs failure when using QuiverKey by @HCookie in #118
• Fixes issue with lead_time indexing from multiple layers by @JamesVarndell in #123
• Not to use gridspec ecCodes key by @sandorkertesz in #124
• fix: Schemas holding state from init by @HCookie in #101
• Adds release notes for version 0.4 by @JamesVarndell in #126
• Improve disabling gridSpec warnings by @sandorkertesz in #125
• Update docs to use intersphinx and improve docs coverage by @JamesVarndell in #121

New Contributors

• @nightowl2626 made their first contribution in #89
• @sandorkertesz made their first contribution in #124

Full Changelog: 0.3.5...0.4.0
Release notes: https://earthkit-plots.readthedocs.io/en/latest/release-notes/version-0.4-updates.html

0.3.5

What's Changed

• Bugfix: prevents striped grid_cells plots when data crosses 0 longitude by @JamesVarndell in #96

Full Changelog: 0.3.4...0.3.5

0.3.4

What's Changed

• Add logos and attributions by @JamesVarndell in #84
• Bugfix/interactive box plots by @leonardung in #68
• 92 memory problem in version 030 by @JamesVarndell in #93

Full Changelog: 0.3.3...0.3.4

0.3.3

What's Changed

• Interpolation and gridspec improvements by @HCookie in #81
• fix Style comparison by @leonardung in #69
• Fix Figure not resolving correctly by @HCookie in #33
• feat/Allow override of dpi in save by @HCookie in #85

New Contributors

• @leonardung made their first contribution in #69

Full Changelog: 0.3.2...0.3.3

Hotfix: Visual artifacts when plotting shapefiles

This hotfix reverts the on-the-fly geometry clipping logic for coastlines, borders, and other shapefiles (issues #75 & #76) and restores full-shapefile plotting to ensure visual correctness. It also fixes a longitude-wrapping bug when extracting data domains that start and end on the same longitude.

Changes

• Instead of attempting to clip shapefile geometries to the current view for performance, we now plot the entire shapefile. This eliminates spurious horizontal lines and artifacts in certain projections, as discussed in #76, but significantly reduces performance.
• When a map's domain wraps around (e.g. –180→+180 or 0→360), exact endpoint values were collapsed (180→–180, 360→0), resulting in tiny slivers instead of the full globe. We now preserve any original +180° or +360° endpoints when re-wrapping longitudes.

What's Changed

• Fixes bugs with extracting data and geometries by @JamesVarndell in #77

Full Changelog: 0.3.1...0.3.2