NEWS

brazilmaps 1.0.0 (2026-08-04)

Preserved the complete public API while making unmatched DTB queries return a schema-stable zero-row data frame.
Documented runtime and maintenance cache behavior, source-provenance gaps and the exact fields in the municipal edition inventory.
Made maintenance downloads validate temporary ZIP or JSON files before replacing cached artifacts, allowed explicit cache/work directories and prevented stale extraction files from entering rebuilds.
Split CI into a fast Ubuntu release check for pushes and pull requests and a comprehensive cross-platform check available on demand.
Added a separate scheduled/manual, rate-limited smoke test for the live IBGE source schema and latest municipal archive.

Added six municipal mesh milestones: 2000, 2001, 2007, 2010, 2023 and 2025. Within each consecutive run with the same municipality count, the latest official edition is bundled.
Municipal files are bundled as topology-preserving compressed TopoJSON. They retain 5% of weighted Visvalingam vertices. Runtime map access is fully local and never downloads data.
Converted current non-municipal maps to topology-preserving compressed TopoJSON without additional simplification, substantially reducing the source package while keeping every spatial object installed locally.
Updated the current hierarchy to 5,571 municipal units, 510 immediate regions and 133 intermediate regions, including Boa Esperança do Norte and the Escada–Ribeirão correction.
Recreated all CRS metadata as SIRGAS 2000 (EPSG:4674) and repaired invalid geometries.
Added a reproducible data-raw/ pipeline with pinned mapshaper, exact official source URLs, checksums and post-serialization validation.

Standardized public functions, arguments, level values and returned spatial columns in lower snake case.
get_brmap() now uses level, filters and output; geographic levels include "municipality", "state", "immediate_region" and related explicit names.
Added brmap_editions(), get_dtb(), get_dtb_levels(), join_brmap() and theme_brmap().
Renamed ambiguous fields to explicit forms such as municipality_code, state_code, immediate_region_code and state_abbreviation.
Filters use intersection (AND), eliminating duplicated rows from multiple conditions.
output = "data.frame" returns attributes without geometry and no longer depends on removed ggplot2 spatial fortify methods.
Previous function names, argument names, PascalCase levels and returned columns remain available through deprecated compatibility paths.
Reimplemented plot_brmap() with geom_sf() and made its theme argument effective. Its join and fill arguments are now data, by and fill_by.
Fixed type preservation in join_brmap() and DTB queries.