Changelog • ucdpR

ucdpR 0.7.1

Added the package author’s ORCID identifier to the structured package metadata so it is displayed and linked by pkgdown.
Standardized the pkgdown navigation with an explicit Articles dropdown that provides direct access to the getting-started guide and research-workflow vignettes.
Added README.Rmd as the canonical README source so the rendered README.md can be regenerated with devtools::build_readme().

Ensured ucdp_to_priogrid() works across supported terra releases whose matrix-coordinate extraction method does not accept an ID argument.
Added ucdp_to_priogrid() to assign georeferenced UCDP observations to the pgid cells produced by the released priogrid package. Custom or frozen grid rasters are supported for reproducible and future-compatible workflows.
Added ucdp_country_crosswalk() for explicit, year-sensitive conversion between Gleditsch-Ward and Correlates of War country identifiers, using peacesciencer or a user-supplied frozen crosswalk.
Added ucdp_to_peacesciencer() to standardize state-year and dyad-year keys, verify their uniqueness, and declare the attributes expected by peacesciencer.
Added an interoperability vignette, help-page examples, and synthetic tests.

Added ucdp_aggregate() for event counts and fatality-bound aggregation at country, PRIO-GRID, conflict, and dyad units with yearly or monthly periods.
Added ucdp_panel() to construct complete analytical panels from raw or aggregated data. Inserted combinations are explicitly identified as zero-filled or missing and never silently presented as observed groups.
Added ucdp_fatalities() to summarize low, best, and high UCDP fatality estimates with completeness diagnostics and configurable missing-value behavior.
Expanded tests, examples, README guidance, and the reproducible-research vignette for aggregation and panel workflows.

Added four focused vignettes covering data access and versioning, spatial workflows, uncertainty and revisions, and reproducible research.
Reworked the getting-started vignette into a concise package overview.
Expanded help-page examples for every exported function, using runnable synthetic data and catalogs wherever the function does not require network access or file downloads.

Added ucdp_as_sf() for provenance-preserving conversion of georeferenced UCDP data to optional sf point objects.
Added ucdp_precision() and ucdp_filter_precision() to describe and filter GED spatial and temporal precision codes.
Added ucdp_version_history() to retrieve official current or historical dataset version histories.
Added ucdp_replication_catalog() and ucdp_replication_download() for publication-specific frozen data, codebooks, version histories, and supplementary files in UCDP’s official replication archive.
Added ucdp_schema() to inspect variable classes, missingness, uniqueness, ranges, examples, and attached provenance.
Added sf as an optional dependency and excluded the default local replication/ download directory from builds and version control.

Excluded the suggested local documentation/ download directory from source builds and version control so saved codebooks do not trigger an R CMD check NOTE.

Added ucdp_metadata() to summarize dataset provenance, dimensions, temporal coverage, and coordinate coverage from imported objects, downloaded paths, or catalog records.
Added ucdp_codebook() to locate and download official UCDP codebooks, including exact-version safeguards and support for documentation variants.
Added ucdp_validate() with structured diagnostics for observation keys, dates, fatality estimates, coordinates, and selected UCDP codes.
New downloads now retain retrieval time and the ucdpR package version in addition to the catalog record, dataset version, and checksum.

Reframed the package documentation around reproducible UCDP research workflows and present and future convenience functions, rather than data downloading alone.
Expanded the README and introductory vignette with a dedicated release- comparison workflow and clearer guidance on package scope and direction.

Corrected the synthetic HTML fixture used to test catalog text extraction. The fixture now spells “Yearly Datasets” correctly while retaining the nested-element structure that guards against dropped letters.

Corrected clean_text(): its former regular expression removed the literal lower-case letters n, r, and t under R’s default regex engine. Whitespace normalization now relies solely on the portable POSIX space class.
Documented the print() and summary() methods separately so their method arguments no longer produce an R CMD check warning.

Fixed catalog titles losing lower-case n, r, and t characters because CSS-aware text extraction treated UCDP’s one-letter markup as hidden.
Restricted URL-based canonical naming to genuine fallback cases, preventing auxiliary datasets from being relabeled as similarly named core datasets.
Applied the same robust text extraction to citation discovery.

Added ucdp_compare_versions() to audit retrospective changes between two downloaded or preloaded UCDP releases. It reports schema changes and added, removed, modified, and unchanged observations, with optional cell-level details and numeric tolerance.

CSV imports now detect type-guessing failures and automatically re-read with full-file inference. This correctly treats occasionally comma-separated UCDP identifier fields as character without slowing every import by default.
Explicit readr controls such as col_types and guess_max remain honored.

Fixed live catalog discovery after UCDP served dataset headings at a different HTML level than the rendered page suggested.
Fixed the multiple-match diagnostic so it includes a stable headline and can be tested reliably.

Initial release.
Added live catalog discovery, search, download, import, citation, and update checking helpers.