Changes in version 0.16.1 (2026-09-05) - Fixed regression #7: legacy SIM/SINASC wrappers and named TABNET filters again accept numeric codes at the start of option labels, translating them to the form's internal values. Exact labels and internal values retain priority; ambiguous numeric prefixes fail explicitly. Offline regression tests cover numeric/text inputs, vectors, leading zeroes and legacy queries. - Registered the RoxygenNote metadata and added package-level documentation that maps the primary discovery, planning, acquisition, validation and analysis entry points while preserving all legacy exports. - Added a monthly, manually dispatchable smoke test for lightweight TABNET and OpenDataSUS metadata checks. It uses bounded requests, downloads no microdata and records actionable schema fingerprints as a workflow artifact. - Expanded deterministic offline tests around the public TABNET and OpenDataSUS discovery contracts. - Revised the package title and description to quote software and health-system names consistently and to expand all domain-specific acronyms requested by CRAN. - Reclassified examples that require network access from \\dontrun{} to \\donttest{}, so they remain user-executable while staying outside routine CRAN example runs. Their request attempts are bounded and temporary upstream failures no longer interrupt checks that explicitly run \\donttest{}. Changes in version 0.16.0 - Added four task-oriented vignettes covering TABNET access, modern OpenDataSUS surveillance, geography and epidemiological analysis, and memory-efficient processing of large microdata files. - Expanded the contemporary schema registry from analysis-critical subsets to the complete fields documented for ESAVI, e-SUS Notifica, PNI doses and the current hospital-occupancy CSV layout. - Added a separate five-minute default download timeout, configurable through options(datasus.download_timeout = ...), while retaining the shorter metadata-request timeout. - Raw DBC/DBF reads now preserve the order requested in colunas, even when the physical DBF stores those fields in a different order. Empty files inside OpenDataSUS ZIP resources now produce an explicit source- unavailable error. - Extended datasus_dicionario() and datasus_padronizar() with curated, officially documented schemas for ESAVI, e-SUS syndrome-gripal, individual PNI doses and COVID-19 hospital occupancy. - Added datasus_validar_esquema() to report missing analysis-critical fields, raw versus standardized columns and incompatible target classes. - Added colunas and normalizar to contemporary wrappers and column selection to opendatasus_ler(). - Added opendatasus_arquivos() to expand resources that publish one or more downloadable file links inside their descriptions. - Fixed e-SUS syndrome-gripal downloads for historical state resources split into multiple lots, while applying n_max globally across all parts. - Added opendatasus_processar() for bounded-memory CSV processing with a callback, configurable chunk size, multipart support and optional schema normalization. Changes in version 0.15.0 - Added esavi() for anonymous, continuously updated vaccination-safety notifications from the e-SUS Notifica ESAVI module. - Added esus_sindrome_gripal() with validated year and state selection for annual mild and moderate influenza-like illness files. - Added pni_doses() for individual monthly National Immunization Program dose records, including dynamic latest-year and latest-month discovery. - Added ocupacao_hospitalar() for annual COVID-19 clinical and intensive care bed occupancy records. - Annual dataset and resource resolution uses live OpenDataSUS metadata and preserves the existing atomic cache, checksum and provenance contracts. Changes in version 0.14.0 - Added calcular_indicador() for grouped rates, proportions, ratios and case-fatality calculations with explicit aggregation and denominator semantics. - Added taxa_mortalidade(), taxa_incidencia(), letalidade() and proporcao() as epidemiological shortcuts. - Added exact Poisson intervals for grouped rates and exact binomial intervals for grouped proportions and case fatality. - Added juntar_populacao() for validated many-to-one joins between event tables and population denominators, including provenance preservation. - Added populacao_padrao() with normalized WHO 2000--2025, Segi and Scandinavian age-standardization weights. Changes in version 0.13.0 - Added datasus_territorios() with an offline IBGE hierarchy covering 5,571 current municipal records, 27 states and five macroregions. - Added normalizar_codigo_ibge(), validar_codigo_ibge() and extrair_codigo_ibge() for seven-digit IBGE and six-digit DATASUS codes. - Added adicionar_territorio() for validated geographic joins and completar_territorios() for missing territory-period combinations. - TABNET and raw-microdata state arguments now also accept two-digit IBGE codes and full state names. - Territorial outputs include source and extraction metadata through datasus_proveniencia(). Changes in version 0.12.0 - Added unified sim() and sinasc() interfaces backed by the declarative TABNET catalog and query engine. - Added explicit municipality, state-specific municipality and region/state query forms through the abrangencia and uf arguments. - Integrated SIM and SINASC with datasus_catalogo(), datasus_opcoes() and datasus_proveniencia(). - Replaced the 15 historical SIM/SINASC implementations with thin compatibility wrappers. Their signatures remain available but now emit a deprecation warning directing users to sim() or sinasc(). Changes in version 0.11.0 - Added calcular_taxa() and exact Poisson intervals with intervalo_taxa(). - Added semana_epidemiologica() and calendario_epidemiologico() using the Sunday-to-Saturday convention followed by Brazilian surveillance calendars. - Added media_movel() with right, centered and left alignment. - Added padronizar_idade() for direct age standardization, Poisson standard errors, optional confidence limits and explicit standard-weight coverage. Changes in version 0.10.0 - Added microdados_catalogo(), microdados_arquivos(), microdados_baixar() and microdados_ler() for record-level DBC files. - Added sim_microdados(), sinasc_microdados() and sih_microdados(). - Added current native DBC/DBF decoding through datasusr, replacing the archived read.dbc dependency. - Added datasus_dicionario() and datasus_padronizar() with conservative field mappings and date conversion for common SIM, SINASC and SIH fields. - Unified the cache root used by OpenDataSUS and raw DATASUS microdata. Changes in version 0.9.0 - Added a generic OpenDataSUS client with catalog search, resource discovery, atomic downloads and CSV/JSON readers. - Added sivep_gripe(), sinan_dengue() and sinan_mpox() for current surveillance microdata. - Added a configurable disk cache with automatic refresh when resource metadata changes, plus URL, timestamp and MD5 provenance metadata. - Added datasus_proveniencia() to retrieve provenance from downloaded files and parsed datasets. Changes in version 0.8.0 - Added pni_imunizacoes() for applied doses and vaccination coverage. - Added siscan() with 15 cervical and breast cancer exam datasets. - Added sisvan() for the two historical nutritional surveillance series. - Added financiamento_sus() for the three financial tables linked by the DATASUS catalog. - Added support for the dhdat/webtabx engine used by PNI and SISCAN, including safe parsing of Google Visualization table definitions. - Made primary form field names dynamic so period selection works across both TABNET engines. - Expanded the local catalog to 118 datasets. Changes in version 0.7.0 - Added populacao_residente() with six census, estimate, projection and retroprojection datasets. - Added sih_morbidade() with eight general and external-causes datasets. - Added sinan() with 46 notifiable diseases and conditions. - Expanded datasus_catalogo() to 96 datasets and added explicit geographic scope metadata. - Added support for TABNET datasets with fixed Brazil/UF forms and for mixed-case endpoint names. Changes in version 0.6.0 - Added a declarative catalog covering SIH/SUS, SIA/SUS and CNES. - Added sih_producao(), sia_producao() and cnes() with a common API for dimensions, measures, periods and named filters. - Added datasus_catalogo() for offline dataset discovery and datasus_opcoes() for live form introspection. - Added support for monthly competencies and Brazilian decimal values in TABNET results. - Added live smoke tests for all three new system families. Changes in version 0.5.0 - Rebuilt the TABNET transport around a shared, retry-aware HTTP client. - Added explicit Latin-1 parsing for current DATASUS responses. - Fixed incorrect evitb10 request endpoints. - Fixed vector handling in the periodo argument. - Made package builds and vignettes independent of live network access. - Added automated tests and continuous integration. Changes in version 0.4.0 (2017-07-25) - 9 new functions related to he SIM's database. - Better filter handling Changes in version 0.1.0 (2017-06-12) - Initial version