Skip to contents

Changelog

Source: NEWS.md

naijR 0.6.1

CRAN release: 2024-10-10

  • Added more function examples and other documentation improvements.
  • Increased compression for internal data.
  • Reflect new home on rOpenSci software repository.

naijR 0.6.0

CRAN release: 2023-08-08

New Features:

  • Added a new dataset states_nigeria for the States of Nigeria.
  • Migrated from RGDAL-based spatial data; the sf package is now the main dependency.
  • map_ng gained a newly documented argument plot, which hitherto was passed internally to maps::map() via ....
  • Added a new function disambiguate_lga, which takes a single-element lgas object for handling the names of Local Government Areas that are shared between separate States. This function enables the (re)setting of the State attribute of such objects; this can be done interactively, where the replicated States’ names are presented for selection by users.

Fixes:

  • Improved on the accuracy of LGA naming in the light of spelling mistakes discovered in the earlier reference document.
  • Unwanted display of dialogs when fixing LGA names on Windows OS was reversed (suggested by Laura DeCicco).

naijR 0.5.2

CRAN release: 2023-06-14

  • Quietened a verbose warning introduced via the soon-to-be-retired spatial data packages.

naijR 0.5.1

CRAN release: 2022-12-02

  • In nested calls with the function fix_region, the lgas constructor function does not warn if there are spelling mistakes, as this turned out to be a bit confusing when it was used. In earlier versions, warnings persisted even after fixes were applied.
  • In carrying out interactive fixes, particularly of LGA spellings, the more familiar and intuitive native Windows messaging and dialog system is used. Works only on Windows machines; on Linux and MacOs, the usual messaging and interaction occurs at the R console.
  • Control the size of map labels with the cex argument (passed on to maps::map.text internally, via ...).
  • Handle instances where the term Abuja is used as a State (which technically it is not) and signal a warning to the user.

naijR 0.5.0

CRAN release: 2022-09-30

New features:

Enhancements:

  • Effectively handle mobile numbers that have common separators in them, namely whitespace, ‘-’ or ‘.’.
  • Repair mobile numbers where poor data entry interchanges zeros (0s) with the letter O (works for both upper and lower case).
  • Repair of mobile numbers now offers optional information for users.
  • Allow the use of factor input when creating objects of class regions.
  • Added a new argument legend.text for map_ng using an idiom that is similar to the one used in base::barplot, thanks to observations made by @VictoriaLatham in issue #27.
  • Simplified the creation of choropleth maps with 2-column data frames; one of the columns is to be a vector of valid States or Local Government Areas, and the other a factor or something coercible to one.

Bug fixes:

  • map_ng accepted arguments that were not data.frames leading to unwieldy errors. It is now made sure to fail early in such cases

Deprecated:

  • Arguments of map_ng - leg.x, leg.y, and leg.orient were marked for deprecation in the next minor release.

naijR 0.4.4

CRAN release: 2022-08-30

Bug fix:

  • fix_mobile fails unexpectedly when only NA is supplied as argument. This causes practical problems when, for example, it encounters a column with only missing values.

naijR 0.4.3

CRAN release: 2022-08-08

  • Addressed a build problem related to CRAN submission.

naijR 0.4.2

  • Improved type checking for mapping functionality and better fidelity.

naijR 0.4.1

  • Enable the exclusion of selected States from a choropleth map (#27).
  • Cleaner output for states and lgas objects.

naijR 0.4.0

CRAN release: 2022-01-28

  • Introduce the ability to ‘manually’ fix names of States or LGAs.
  • Update the documentation with a new vignette.

naijR 0.3.4

CRAN release: 2021-12-06

  • Fixed package-wide misuse of the word Nasarawa.

naijR 0.3.3

CRAN release: 2021-11-05

  • Fixed repetitions in the output when multiple LGAs’ spellings are corrected.

naijR 0.3.2

CRAN release: 2021-10-20

  • Improved on print methods

naijR 0.3.1

CRAN release: 2021-09-30

  • Fixed a bug that affected the proper rendering of LGA-level maps for some of the States. The approach used was to simply filter the entire data when requiring a State map, so as to reduce name clashes that occurred from synonyms amongst some of the LGAs and/or States.
  • Enabled the fine-tuning of creation of lgas objects in the event that the argument provided is the name of an LGA that is synonymous with it’s State (argument strict).

naijR 0.3.0

CRAN release: 2021-09-08

  • Provide new methods for the S3 generics head and tail to work with objects that inherit from class regions.
  • The S3 constructors states and lgas gain a logical argument warn to control whether or not they issue a warning when an input string does not contain an actual State/LGA.
  • General improvement of the formatting of output to enhance the user experience.
  • Fixed a bug that prevented the loading of LGAs from the internal data when the package is not attached to the search path i.e. invocation with naijR::lgas() was producing an error.

naijR 0.2.2

CRAN release: 2021-04-30

  • Export S3 generic fix_region.

naijR 0.2.1

CRAN release: 2021-04-25

  • Fixed incorrect URLs, as noted by CRAN
  • Edits to output message

naijR 0.2.0

  • Added a new function is_lga, which checks an object for Local Government Areas.
  • Ignore, with a warning, the check for is_state when the object checked is not of type character.
  • Draw maps up to LGA level

naijR 0.1.5

  • Built new package website.

naijR 0.1.4

CRAN release: 2020-08-25

  • Suppress deprecation warning for is_state when it is called internally by package function; displayed only when function is called directly.

naijR 0.1.3

CRAN release: 2020-08-03

  • Added a NEWS.md file to track changes to the package.
  • Recognise abbreviations of ‘Federal Capital Territory’ i.e. FCT.
  • Disable error-check on character type for is_state so it can be used more effectively for functional programming constructs.