NEWS

dtrackr 0.4.6 (2024-10-21)

fixed warnings caused by additional deprecated dplyr functions.
More informative messages when syntactic errors in exclude_all and include_any input
fixed regression bug in pivot_wider
documentation updates.
improved ranking of nodes and branching in groups.

Updating release following dplyr version 1.1.0 an changes in new versions of tidyr
More use of @inheritDotParams to decouple from upstream APIs
fix for #33
Fix for RSVG scaling inconsistency / rotation issues.

Updating release in preparation for dplyr version 1.1.0
Fix problems detected with URLs and DOI.
Updated URL: https://terminological.github.io/dtrackr to https://terminological.github.io/dtrackr/ in readme README.md (fixed missing trailing slash)
Manually fixed auto-generated codecov badge.
Removed project level Zenodo DOI as triggering a invalid DOI from CRAN
Updated maintainer email to longer term stable email address. URLs in
DESCRIPTION updated to point to github source.
There are some spelling issues that arise from documentation inherited from dplyrand tidyr equivalent functions, that would need to be fixed upstream.
Documentation inherited from dplyr may be inconsistent until dplyr release.

updates following JOSS review: changes in documentation have resulted in additional Suggests dependencies on CRAN packages.
bug fix for group_by not checking pause status before warning that pause required.
documentation improvements - full review of all documentation. Additional context to function reference to draw attention to supported dplyr operations
added nest_join.tracked_df S3 method, added slice*.tracked_df functions, and missing set operations functions.
re-factored mutate style operations, regression tested
automated testing for new methods, and full testing for slice, set ops, and mutate style functions with new example code.
expanded test coverage ( covr::package_coverage(type="all" ): dtrackr Coverage: 83.60%, R/dot.R: 78.12%, R/dtrackr.R: 84.08%) mostly through examples
improved error handling and reporting in mutate functions.
bug #25 & #26 fixes (see github)
Logo and documentation fixes and improvements.
outstanding issues:
re-use of documentation results in examples tests running multiple times.

second round fixes for CRAN. Improved examples in documentation.
Print and plot S3 methods for history graph.
Vignette fixes.
Non breaking changes to API so bumping to new minor version for CRAN submission.

documentation updates & JOSS paper
fix for fix for multiple names join columns bug.
ungroup() change to fit with dplyr::ungroup API.
fixed url redirects, DESCRIPTION file issues and switched to LICENSE file template, for CRAN submission.

new feature of subgroup counts for determining size of subgroups in grouped (or un-grouped) flowchart.
fix for multiple names join columns bug.
bug fixes and extended github workflows.

minor bug fixes for grouping
support for tagging pipeline with a piece of data for later retrieval to support e.g. for counts within abstract
support for tidyselect syntax in group by

track excluded items with debugging info
allow across syntax in most situations (except group_by)
support pausing and un-pausing of the dataframe tracking
preventing massive group_by groups if large numbers of very small groups are selected (e.g. part of a group_by() %>% mutate())
switch warnings to rlang::warn etc.

Initial package features complete
dplyr like for like functionality supporting - add_count, add_tally, anti_join, arrange, bind_rows, comment, distinct, exclude_all, filter, full_join, group_by, group_modify, inner_join, left_join, mutate, pivot_longer, pivot_wider, relocate, rename, rename_with, right_join, select, semi_join, summarise, transmute, ungroup
add a comment into the history (stratified by grouping),
add a status into the history (stratified by grouping),
apply exclusion and inclusion filters including stratified commentary
merge histories of joining data frames
export history as dot graphviz graph
render history graph to SVG, PNG, PDF, DOT or PS formats