Changes in version 1.0.0 (2026-05-16)                  

Breaking changes

  - tidy_bunch(): the merge parameter has been renamed to combine to
    avoid shadowing base::merge(). Update any existing calls
    accordingly.
  - tidy_indice() is deprecated in favour of the correctly spelled
    tidy_index(). It still works but will emit a deprecation warning. It
    will be removed in a future release.

New features

  - tidy_index() and tidy_bunch() now return tibbles instead of plain
    data frames.
  - tidy_bunch() now returns a named list (named after each indicator
    file) when combine = FALSE.
  - tidy_bunch() now supports .xls files in addition to .csv and .xlsx.
  - Both functions now emit informative error messages for missing
    files, missing directories, unsupported file extensions, and empty
    directories.
  - Year coercion to numeric now emits a warning if any values cannot be
    converted.

Internal changes

  - Migrate CI from Travis/AppVeyor to GitHub Actions.
  - Drop support for R < 4.1.0.
  - Upgrade to testthat 3rd edition with expanded test coverage.
  - Replace readr::read_csv() with base read.csv() and
    tidyr::pivot_longer() with base reshape(), removing both readr and
    tidyr from dependencies. Package now depends only on readxl and
    tibble.
  - pkgdown site upgraded to Bootstrap 5.

                 Changes in version 0.1.1 (2020-02-04)                  

  - Fix a bug in merging list of data frames in tidy_bunch().

                 Changes in version 0.1.0 (2019-12-06)                  

  - First release, with tidy_indice() and tidy_bunch() functions.