Skip to contents

Changelog

Source: NEWS.md

ggcircular 0.2.0

New features

  • Integrated the directional-dependence displays into ggcircular.
  • Added circular-linear helpers for density estimation, phase loom displays, circular topographies, statistical orbits and bootstrap orbit intervals.
  • Added circular-circular helpers for toroidal density estimation, toroidal topographies, toroidal flow displays, conditional ridges and bootstrap ridge intervals.
  • Added directional diagnostic scenarios, diagnostic atlases, classical unfolded comparisons, marginal support diagnostics, smoothing selection and independence lineups.
  • Added simulation helpers for cylindrical and toroidal examples.
  • Added experimental plotly-based 3D companions for circular and toroidal topographies.
  • Added near-tie diagnostics for conditional toroidal ridges while preserving the first-grid-maximum selection rule.
  • Corrected signed-angle wrappers so all observations contribute to toroidal topographies, flows and ridges.
  • Replaced seam-sensitive linear ridge bootstrap intervals with pointwise circular intervals.
  • Expanded the README and vignettes with executable conditional-display examples and practical marginal-support guidance.

API

  • The existing public API remains available.
  • New ggplot2 layers use the standard ggcircular convention aes(x = theta, y = response) or aes(x = theta, y = phi).
  • High-level plot_*() helpers are retained as shortcuts built from the new ggplot2 layers.
  • Experimental 3D helpers return plotly htmlwidgets and keep plotly in Suggests.
  • Prototype-only integration stubs are not included.

ggcircular 0.1.0

CRAN release: 2026-06-04

New features

  • Added rose diagrams, circular density estimates, mean direction layers, circular scales, circular coordinates, movement helpers and simulated datasets.
  • Added circular summaries, confidence intervals, Rayleigh and Watson-Williams helpers, and theoretical circular distribution overlays.

Experimental

  • Added angular model diagnostics and optional helpers for CircularRegression.
  • Added exploratory finite mixtures of von Mises distributions.
  • Added optional helpers for momentuHMM, spherical summaries and posterior circular draws.
  • Added repository support files for contributions, support and conduct.

Documentation

  • Added a README with badges, examples and a package logo.
  • Added pkgdown reference pages and articles for the main workflows.
  • Added CRAN readiness notes to the README, validation vignette and cran-comments.md.

Validation

  • Added testthat coverage for angle utilities, summaries, ggplot2 layers, diagnostics, movement helpers, circular tests, spherical/posterior helpers and von Mises mixtures.
  • Added strict momentuHMM alignment checks to prevent silent truncation.
  • Added a strict hard-dependency CI profile with _R_CHECK_FORCE_SUGGESTS_ set to false.

Known limitations

  • Advanced model diagnostics, von Mises mixtures and optional ecosystem integrations are experimental.
  • Automatic circular-density smoothing is heuristic.
  • Further validation is planned for the R Journal article workflow.