Skip to contents

Changelog

Source: NEWS.md

badata 0.2.0

  • New data delivery from BA (Auftragsnummer 408036) extends coverage: employees_by_workplace/employees_by_residence now span 2013-2025 (previously 2016-2021), and jobs/unemployed_total/ unemployed_foreigners now span 2012-2025 (previously 2012-2021).

  • Fixed two parsing bugs surfaced by a change in BA’s export format:

    • the “darunter” (foreigners/women) sub-columns were colliding into identical names across the two employee types, silently dropping the foreigners/women breakdown for one of them
    • the un-labeled “total” sub-column was being dropped entirely instead of set to "total" in the group column of employees_by_workplace/employees_by_residence

  • Fixed a pre-existing bug, present since the original release, where region_occupation_codes.R picked up the source sheet’s trailing metadata line (“Erstellungsdatum: …”) as a bogus entry in region_codes.

  • Fixed occupational_group_codes’s “ohne Angabe zum Zielberuf” catch-all category, which previously had the same string duplicated into both code and name. It now has code = NA and the correct name, so a left_join() against the NA occupational_group values in the main tables now attaches the label correctly.

  • Hanau split off from Main-Kinzig-Kreis as an independent city (region code 06415), retroactively restated across all historical years per BA’s “Gebietsstand Juli 2026” convention. Anyone comparing region 06435 figures against pre-0.2.0 data will see a level shift, since it now excludes Hanau’s share.

  • 4 military occupational codes (011-014, e.g. “Offiziere”) are no longer reported in the employees export and are absent from employees_by_workplace/employees_by_residence from this release on.

  • Remaining differences in overlapping years (roughly 0.3-1.5% of matched cells) reflect BA’s own disclosed data-revision policy, not a parsing issue.

  • Konstantin Wandel joins as a package author.