Changelog • lineager

lineager (development version)

Initial release.

lg_start() / lg_end() — session lifecycle. lg_start() initialises a clean session store; lg_end() prints a summary and marks the session inactive. The store remains queryable until the next lg_start().
lg_tag() — entry point. Assigns a unique lineage ID (.__lid__) to every row. For datasets with a USUBJID column, the ID embeds the subject identifier for human readability (DM_0001_01-042). For general datasets, a zero-padded sequence is used (patients_000001).
lg_filter() — tracked filter with mandatory reason. Every row removal is documented and captured in the session exclusion registry. Optional reason_code and population arguments enrich the record.
lg_derive() — tracked dplyr::mutate() with mandatory description. Every variable derivation is recorded in the operation log.
lg_join() — tracked join (left, inner, full, right). Preserves .__lid__ from x; adds .__lid_y__ to record which rows of y contributed, enabling bilateral tracing.

lg_population() — register a population or cohort flag definition (SAFFL, ITTFL, PPROTFL, or any custom flag). Records inclusion/exclusion criteria, plain-English definition, and automatic counts.
lg_spec() — document a source-to-analysis variable derivation. Links output variables back to their source variables and datasets.

lg_trace() — trace any row’s complete lineage journey: which datasets it appears in, which operations it passed through, and any exclusion records with documented reasons.
lg_exclusions() — retrieve the full exclusion registry as a data.frame, filterable by population or dataset.
lg_disposition() — CONSORT-style subject disposition summary, grouped by reason, population, or dataset.
lg_operations() — full pipeline operation log as a data.frame.

lg_lineage() — build a lineage graph of the complete pipeline: source datasets, all derive/join/filter operations, and exclusion branches, as a Graphviz DOT string.
lg_plot() — render the lineage graph inline (via DiagrammeR if installed) or write the DOT source to a file.

lg_report() — generate a self-contained HTML provenance report covering dataset inventory, subject disposition, population flag definitions, variable derivations, operation log, and exclusion listing. For CDISC users, the output aligns with CDISC Reviewer’s Guide content requirements.