Skip to content

Releases: ipea/geobr

v2.0.1

Choose a tag to compare

@rafapereirabr rafapereirabr released this 23 Jun 12:39

Bug fixes

  • Moves the {arrow} package from "Suggests" to "Imports".
  • Fix bug in read_immediate_region() which was hardcoded to read state 11 by mistake. Closes #436.
  • The function remove_islands() now correctly drops the arquipelago de Trindade e Martim Vaz
  • Fix documentation of arguments year and date.
  • Make the internal function that downloads geobr metadata more robust and removes the dependency from piggyback
  • More informative message when a municipality does not have neighborhood data. Closes #424.

v2.0.0

Choose a tag to compare

@rafapereirabr rafapereirabr released this 20 May 12:59

New functions

  • read_favela() with data of favelas and urban communities (source: IBGE) Closes #387.
  • read_polling_places() with data of polling places (source: TSE) Closes #184 and #242.
  • read_quilombola_lands() with data of officially recognized quilombola lands (source: INCRA) Closes #242.
  • remove_islands() to remove islands from Brazil. Closes #412.

Breaking changes

  • The year and date arguments can no longer be NULL; they must be explicitly
    specified. This change is intentional and is meant to encourage users to be more
    mindful of historical changes in the data.
  • The geom column has been renamed to geometry for consistency
  • The read_health_region() has been completely rewritten to allow users return
    more detailed output if needed
  • Functions like read_schools() and read_health_facilities() now use a
    combination of official spatial coordinates and coordinates found using the
    {geocodebr} package to improve spatial
    accuracy. See documentation of these functions.
  • The function lookup_muni() now has a year parameter. Closes #401.
  • The function and data read_comparable_areas() will be going under major
    changes. For now, this function is temporarily suspended.
  • The only year available so far for the functions read_urban_concentrations()
    and read_pop_arrangements()is 2010, and not 2015.

Major changes

  • Data files are now saved in .parquet. This improved performance to download
    and to read files, and allow integration with ducdkDB via {duckspatial} and with
    Arrow via {gearrow}. Closes #290
  • Most functions have a new argument output, which allow users to choose the
    output format. "sf" returns an sf to memory (default), "duckdb" returns a
    lazy spatial table backed by DuckDB via the duckspatial package, and "arrow"
    returns an Arrow dataset. Both "duckdb" and "arrow" support out-of-memory
    processing of large data sets.
  • All functions have a new argument verbose. If TRUE (the default), the
    function prints informative messages and shows download progress bar. If FALSE,
    the function is silent. Closes #400.
  • The function list_geobr() now has a boolean argument wide, so users can
    choose whether the output should be presented in wide or long format.
  • The function lookup_muni() now uses probabilistic match to find municipality
    names that users might input with typos. Closes #406.
  • The following functions now include the column code_state to allow users
    to filter the data directly in the function call: read_indigenous_land(),
    read_metro_area(), read_pop_arrangements() and read_urban_concentrations().
  • The following functions now include the column code_muni to allow users
    to filter the data directly in the function call: read_disaster_risk_area(),
    read_health_facilities(), read_neighborhood(), read_statistical_grid() and
    read_schools().

Minor changes

  • Several data fixes and data updates, addressing the following issues: 182, 247,
    249, 250, 267, 333, 340, 361, 369, 379, 384, 388, 389, 390, 391, 393, 404, 407.

New co-author

  • Rogerio Barbosa

New contributors

  • Cecilia do Lago
  • Arthur Bazolli
  • Filipe Cavalcanti
  • Lucas Gelape
  • Rafael Lopes
  • Vinicius Oike

New funding / institutional support

  • Instituto Todos pela Saúde (ITpS)

v1.9.1

Choose a tag to compare

@rafapereirabr rafapereirabr released this 07 Sep 00:40

Minor changes

  • The read_municipality() has a new parameter keep_areas_operacionais, which allows users to control wether the data should keep the polygons of Lagoas dos Patos and Lagoa Mirim in the State of Rio Grande do Sul (considered as areas estaduais operacionais). The default FALSE drops these two polygons. Closes #176.
  • Functions now include a cache parameter that allows users to decide whehter to keep files in cache or to force downloading them again. At the moment, files are only cached during the R session, but this is a step towards a future version of {geobr} when files will be based on permanent caching.
  • Now using curl::multi_download() to download files in parallel
  • Removed dependency on the {httr} package
  • {geobr} now imports {fs} to use robust cross-platform file system operations
  • Simplified and streamlined internal functions

v1.9.0

Choose a tag to compare

@rafapereirabr rafapereirabr released this 18 Apr 22:36

Major changes

  • Function read_health_facilities() now has a new parameter date, which will allow users to access data for different dates of reference. The plan is to have at least one update of this data set per year. Closes #334.
  • Functions read_urban_area() and read_metro_area() now have a new parameter code_state, which will allow users to filter selected states. Closes #338

Bug fix

  • Using data.table::rbindlist() to rind data was throwing errors when some observations were of class POLYGONS and others were MULTIPOLYGONS. This has now been replaced with dplyr::bind_rows() at a very small performance penalty. Closes #346.

New data

  • schools for 2023
  • health facilities for 202303
  • census tracts for 2020 and 2022

v1.8.2

Choose a tag to compare

@rafapereirabr rafapereirabr released this 09 Jan 13:06

CRAN request

  • Fixed issue to make sure geobr uses suggested packages conditionally

Minor changes

  • Fixed non-ASCII characters in data geobr::grid_state_correspondence_table()

v1.8.1

Choose a tag to compare

@rafapereirabr rafapereirabr released this 21 Sep 12:50

CRAN request

  • geobr now uses suggested packages conditionally

v1.8.0

Choose a tag to compare

@rafapereirabr rafapereirabr released this 09 Sep 22:28

New function

  • read_capitals() to download either a spatial sf object with the location of the municipal seats (sede dos municipios) of state capitals, or a data.frame with the names of codes of state capitals. Closes #243

Minor changes

  • Update intro vignette to show how to use geobr together with the new censobr sister package.

Bug fixes

  • fixed bug from conflict between sf and data.table that was messing with plot extent. Closes #284.
  • fixed bug from conflicts between plotly and data.table. Closes #279.
  • fixed bug in cep_to_state() function. Closes #317.
  • fixed bug in progress bar. Closes #154.
  • The lookup_muni() and download_metadata() functions are now more robust to internet connection failures.

v1.7.0

Choose a tag to compare

@rafapereirabr rafapereirabr released this 12 Aug 22:12

Data release

v1.6.5

Choose a tag to compare

@rafapereirabr rafapereirabr released this 04 Jan 12:36

geobr v1.6.5

Minor changes

  • Remove packages cruland readr from geobr dependencies.
  • geobr now fails gracefully when server connection times out. Closed #259.

Bug fixes

  • Fixed check_connection() that was failing on Linux. Closed #269.

v1.6.4

Choose a tag to compare

@rafapereirabr rafapereirabr released this 04 Jan 02:00

bug fixes

  • Fixed bug crashing on Solaris.