GisBR
A QGIS plugin that brings Brazil's official spatial datasets straight into the Processing Toolbox.
The Problem
Anyone doing spatial analysis in Brazil knows the routine: building a municipal map means browsing IBGE's FTP server, finding the boundary files for the right year, downloading the archive, extracting it, and only then loading it into QGIS. The geobr package, from IPEA, solved this elegantly for R and Python users — one function call returns a ready-to-use layer — but left out most GIS professionals, who work directly in QGIS and don't write code.
The same applies to diagnosing a municipality for a Master Plan: gathering official transport, sanitation, demographic, and environmental datasets means manually querying dozens of different WFS servers, ArcGIS REST endpoints, and download portals, each with its own format and reference system.
The Solution — Project Evolution
Approved in the official QGIS plugin repository. Each geobr geography becomes a Processing algorithm: 55 in total, covering administrative boundaries, census data, urban and metropolitan areas, environmental themes, and sectoral data such as health and education, plus the join_censo algorithm, which joins census tracts with censobr's demographic tables. Design decision: zero external dependencies, only the native QGIS/Qt API and Python's standard library.
Downloads the road network of any municipality directly from OpenStreetMap via the Overpass API, structured as links and nodes with minimal topology — the same logic behind osm2gmns, by professor Xuesong Zhou (Arizona State University), applied inside QGIS. Also fixed a timeout value that was never applied to Overpass requests. First release published without the "experimental" label.
Explicit support for QGIS 4.x, plus a fix for a PyQt6 enum-scoping bug that took down the municipality catalog and the Overpass network error check.
SSL certificate support, until then limited to the IPEA downloader, moved into a single module (core/ssl_support.py) and was extended to every connector in the plugin — WFS, ArcGIS REST, Overpass, and the satellite basemap — fixing certificate chain failures on fresh Windows/OSGeo4W installs when querying sources like SICAR, ANA/SNIRH, INDE, SGB, and IBAMA.
The diagnostic's source catalog gained six new public datasets: geological risk sectors (SGB/CPRM), mining claims (SIGMINE), groundwater wells (SIAGAS), IBGE's BC250 and urbanised-areas layers, and the BDIA soil, geology and vegetation set (v0.5.0). The following release added a SIGEF connector that reads certified land parcels from a manually downloaded file, since INCRA's export service requires a gov.br login (v0.6.0).
v0.7.0 added an OpenStreetMap POI engine that downloads buildings, businesses and station-like points for the urban axis, plus a GMNS export algorithm for demand models. v0.8.0 let the diagnostic attach Census tables via censobr directly to the census tract layer.
Ninth diagnostic axis, "Agriculture and Livestock," with MapBiomas land cover and land use data plus IBGE agricultural data (PAM and the 2017 Census of Agriculture).
The road network was rebuilt from OpenStreetMap's real topology: ways are split into arcs at shared nodes, and the vehicle and pedestrian networks are now generated separately. A new layer collects connectivity problems (islands, dangling ends, one-way traps) classified by severity, replacing the minimal topology from v0.3.2, which lost junctions in the middle of a way.
First 1.0 release: the diagnostic gained a Metropolitan Region mode, running over every municipality in a metro area instead of a single one, covering the 84 metropolitan regions in IBGE's catalog.
Key Features
57 Processing algorithms: every official geography (country, states, municipalities, census tracts, biomes, conservation units, indigenous lands, favelas, polling places, and more) becomes an algorithm with year, state/IBGE code filtering, and optional simplified geometry.
Master Plan diagnosis: gathers 54 official sources organized into 9 axes, already clipped to the municipality's boundary, into a single GeoPackage.
Metropolitan Region mode: runs the same diagnosis over every municipality in a metropolitan region, covering the 84 RMs in IBGE's catalog.
OpenStreetMap road networks: downloads any municipality's roads via Overpass, built from OSM's real topology, with ways split into arcs at shared nodes. Vehicle and pedestrian networks are generated separately, and a dedicated layer collects connectivity problems (islands, dangling ends, one-way traps) with severity, in SIRGAS 2000 (EPSG:4674).
Local cache with a GitHub mirror: downloads are cached on disk; if IPEA's server is down, the plugin fetches the same files from a GitHub mirror without interrupting the workflow.
Embedded SSL certificates: recognizes the trust chains of IPEA, SICAR, ANA/SNIRH, INDE, SGB, IBAMA, Overpass, and the satellite basemap, fixing failures on fresh Windows installs without relaxing any security check.
Zero external dependencies: runs on only the native QGIS/Qt API and Python's standard library — no geopandas, pandas, or requests — avoiding conflicts with other plugins and installation barriers for less technical users.
Technology Stack
Python PyQGIS QGIS Processing Framework Overpass API GeoPackage / SQLite Parquet / GDAL
Impact
GisBR removes the manual search, download, and conversion steps for anyone working in QGIS, covering dozens of administrative, census, environmental, and sectoral geographies, plus OpenStreetMap road networks — all without requiring any programming knowledge. The plugin also runs the municipal diagnosis for drafting and revising Master Plans, for a single municipality or a whole metropolitan region, gathering the official datasets already clipped to the area's boundary into a single GeoPackage.