NEWS

dtlg 0.1.2

Added and exported aesi_heading_msg(), which provides the shared default heading used for adverse events of special interest in the native and {tern} AET01 tables.

as_dtlg_table() now uses where(is.character) when selecting columns, removing a tidyselect deprecation warning.
Reinstated the internal %||% operator and documented its explicit use to avoid a false-positive dead-code warning from {goodpractice}.
Replaced vctrs::vec_recycle_common() with a focused base R implementation and removed {vctrs} from package dependencies. Existing summary_table() argument-recycling behaviour is preserved.

Extracted input validation and table-layout helpers from AET01_table() and tern_AET01_table().
Extracted descriptive-statistics calculation and validation helpers from calc_desc().
Removed unused column-relevel helpers and moved format-matching utilities from package code into the test suite.
Expanded tests for AET01 and AET02 tables, {tern} table layouts, calc_stats(), as_dtlg_table(), printing, labels, package availability, contingency-table expansion, and vector recycling.
Strengthened existing tests with stricter type, structure, formatting, and error assertions.

Added the rOpenSci pkgcheck workflow.
Refined the lintr configuration and added targeted exclusions for established API names.

AET01_table(): the row label for AE-related withdrawals has been shortened from "Total number of patients withdrawn from study due to an AE" to "Total number of patients withdrawn due to an AE", for consistency with the other terse event labels in the same table. Code that matches this label by string equality (e.g. snapshot tests) will need to be updated.

The README and the "Getting started" article now set a wider print width and display the full AET01 table rather than truncating row labels.

summary_table(), summary_table_by() and summary_table_by_targets() gain an inc_missing argument controlling how the "Missing" row is rendered for numeric targets: TRUE (default) always shows it, NA shows it only when missing values are present, and FALSE suppresses it. Plumbed through to calc_desc() and calc_stats().
summary_table_by() and summary_table_by_targets() gain a sep argument to set the separator between the heading and group labels when grouping by more than one variable in rows_by (defaults to ".").

calc_desc(): the n row now reports the count of non-missing values rather than the total row count (.N), so n and the Missing row are now consistent.
summary_table_by_targets(): removed the spurious "target needs to be length 2" message; multiple target variables are now properly supported and covered by tests.

Adopted lintr with default linters across the package and added a lint.yaml GitHub Actions workflow.
Added a code-coverage workflow that posts a coverage summary as a comment on pull requests, gated on pull_request events to avoid duplicate runs.
Expanded the test suite -- new tests for AET01_table(), the rounding helpers, tern_summary_table(), and additional summary_table*() cases (multiple targets, missing-value handling) -- raising overall coverage.

Reorganised maintainership and contact details: maintainer e-mails moved to the acuityanalytics.com domain and copyright holder updated to Acuity Analytics.
Documentation cleanup for dt_count(), dt_summarise() and the summary_table*() family; minor README and benchmarks article fixes.