Skip to contents

Changelog

Source: NEWS.md

tidycreel 1.4.0 (2026-04-23)

Quality, testing, and release readiness

  • Closed the priority rOpenSci blocker set for the current release line: named condition classes at the key cli_abort() sites, formal lifecycle badges on experimental APIs, a valid inst/CITATION, and removal of the scales dependency from the package surface.
  • Demoted lubridate from Imports to Suggests and added runtime install guards at user-facing schedule entry points.
  • Threaded rlang::caller_env() through the top-level bus-route estimator internals and relocated get_site_contributions() into the estimation layer to tighten call-frame quality and layering.
  • Added @family tags across the exported surface so the pkgdown reference is grouped by workflow topic rather than a flat function list.
  • Added snapshot regression coverage for print.creel_design(), print.creel_estimates_mor(), and print.creel_schedule().
  • Added quickcheck-based property tests and generator helpers covering the highest-value implemented invariants: INV-01, INV-02, INV-03, INV-04, and INV-06.
  • Added a CI-backed coverage gate with a documented local baseline of 86.27%, Codecov configuration, and a project target of 85%.

tidycreel 1.3.0

New features

  • estimate_catch_rate() now accepts estimator = "mortr" for truncated mean-of-ratios (MORtr), which applies truncate_at as a mandatory threshold and labels the method "mean-of-ratios-truncated-cpue".
  • estimate_catch_rate() gains a targeted argument (default TRUE). Setting targeted = FALSE excludes zero-catch trips before MOR/MORtr estimation for incidental species workflows.
  • power_creel() provides a unified tidy entry point for pre-survey sample-size planning, wrapping creel_n_effort(), creel_n_cpue(), and creel_power() into a single consistent interface with mode = "effort_n", "cpue_n", or "power".
  • compare_designs() compares multiple survey designs side by side from a named list of creel_estimates objects. An autoplot() method renders a forest plot of point estimates with confidence intervals.
  • as_hybrid_svydesign() constructs a hybrid access + roving survey design from combined access-point and roving-route count data.
  • compare_variance() computes Taylor linearization vs. replicate (bootstrap or jackknife) standard errors side-by-side for any creel_estimates object.
  • adjust_nonresponse() applies nonresponse weighting to a creel_design and records per-stratum diagnostics.
  • est_effort_camera() adds ratio-calibrated camera/time-lapse effort indexing.
  • est_length_distribution() adds weighted catch-at-length / size-structure estimation from attached length data.
  • autoplot.creel_length_distribution() adds a plotting surface for weighted size-structure estimates.
  • theme_creel() and creel_palette() add package-standard plot styling.

Data validation and cleaning

  • validate_creel_data() adds field-level schema validation for creel inputs.
  • standardize_species() adds canonical species-code standardisation helpers.
  • validation_report() adds formatted validation summaries that can be exported alongside other report-ready outputs.
  • creel_counts_toy and creel_interviews_toy are now bundled example datasets for examples, tests, and documentation.

Documentation and reporting

  • Added a glossary vignette for package terminology and workflow language.
  • Added a survey design toolbox vignette covering planning and pre-season tools.
  • Added a flexdashboard report template scaffold under inst/rmarkdown/templates/creel-dashboard/.
  • Expanded pkgdown/reference discoverability for the newer estimation, visualisation, and reporting surfaces.
  • The full pkgdown site now rebuilds cleanly after normalizing older vignette header/title inconsistencies.

Improvements

  • plot_design() now supports multi-strata designs.
  • Main estimator autoplot() methods now support opt-in theme = "creel" styling without changing default behavior.
  • Single-PSU strata produce a structured, actionable error instead of an opaque survey:::onestrat message.
  • Fixed a bug in the aerial-glmm vignette downstream estimation chunk where example_aerial_interviews was paired with the wrong design object.

Dependencies

  • ggplot2 added to Imports to support the autoplot.* methods.
  • flexdashboard added to Suggests for the optional report template.

Tests

  • Expanded test coverage for the newer estimation, validation, plotting, and reporting surfaces shipped through the current 1.3.0 development line.

tidycreel 1.2.0 (2026-04-08)

New features

  • summary.creel_estimates() converts any estimate object to a creel_summary with human-readable column names (Estimate, SE, CI Lower, CI Upper, N). Includes print.creel_summary() and as.data.frame.creel_summary() methods. Works for effort, CPUE, harvest rate, total catch, and grouped variants.

  • flag_outliers() identifies extreme values in a numeric column using Tukey’s IQR fence (k = 1.5 default). Returns the input data frame with is_outlier, outlier_reason, fence_low, and fence_high columns appended, and emits a cli summary of flagged rows. Handles n < 4, empty input, and zero-row data frames gracefully.

  • ggplot2::autoplot.creel_estimates() produces a point-and-errorbar plot from any creel_estimates object. Ungrouped estimates show a single point with confidence interval; grouped estimates show one point per group level, colour-coded.

  • ggplot2::autoplot.creel_schedule() produces a monthly tile calendar from a creel_schedule object. Sampled dates are coloured by day type (weekday blue / weekend red); unsampled dates are shown in grey. Multiple months are displayed as vertically stacked facet panels.

Improvements

  • Single-PSU strata now produce a structured, actionable error instead of an opaque survey:::onestrat message. The error names the problematic stratum and suggests increasing the sampling rate or combining sparse strata.

  • Fixed a bug in the aerial-glmm vignette downstream estimation chunk where example_aerial_interviews was paired with the GLMM design (built from example_aerial_glmm_counts). The chunk now uses the correct matching dataset (example_aerial_counts + example_aerial_interviews).

Dependencies

  • ggplot2 added to Imports to support the new autoplot.* methods.

tidycreel 1.1.0 (2026-04-02)

New features

Documentation

  • GitHub issue templates now use structured forms with blank_issues_enabled: false, routing how-to questions to GitHub Discussions to keep answers searchable for all users.

  • CONTRIBUTING.md has been rewritten with current workflow guidance, contribution types, and community norms for the v1.x release line.

tidycreel 1.0.0 (2026-03-31)

  • Launched the pkgdown documentation site at https://chrischizinski.github.io/tidycreel with a custom Bootstrap 5 theme, full function reference index (46 exports + 15 datasets), and a workflow-driven navbar.

  • Added a GitHub Actions CI/CD workflow to deploy the pkgdown site automatically on every push to main.