- Implement a cache for API calls (including file download) when dataset version is specified. The functions will reload from the cache automatically the second time. (#112, #135, by @mtmorgan)
- Improve recommendation for rdata loading (#107, #127)
get_file_by_*()can now return the download URL to be used in external functions or programs, useful for large files (#128, implemented in #129 @JBGruber and @kuriwaki)- Removes remote resource from vignette and move them to ghactions (#131)
- Update expired token (#123)
- Allow for downloading from draft datasets (#115, @Danny-dK)
- Fix add dataset bug introduced in a recent version (#116)
- Add progress bar for all downloads (#108)
- Minor documentation improvements (#64, #107)
- Faster method for detecting ingest (#113) and robustness to ingested files without a metadata file due to errors (#80)
- No longer relies on
foreign(#34)
- Change maintainer to @kuriwaki
Core Features
- Adapt to Dataverse 5.4.1 (#94 @kuriwaki)
Testing
- Doesn't rely on CRAN for daily tests of API functions (#96)
- Upgrade testthat to 3.0 edition (#97)
- Move to GitHub Actions, and away from Travis-CI. (#98) Three GitHub Actions are defined for testing:
- thorough testing related to PRs and the main branch,
- straight-forward tests for each push to the dev branch, and
- straight-forward tests run daily at 3am
- More verbose error messages returned on httr failure. (#31 @EdJeeOnGithub)
- Avoids tests in R Check if the dataverse server is not available (#77)
- Avoids potential mixup of dataset ordering within a retrieved dataverse (#83)
New Methods
- Add new
get_dataframe_*()methods (#48, #66)
Small updates
- Make filter queries (fq) work in
dataverse_search(#36 @adam3smith) - Update maintainer to Will Beasley ([email protected]) (#38)
- More robust file retrieval (#39 @kuriwaki)
- Tests use https://demo.dataverse.org/dataverse/dataverse-client-r/. (#40)
- Fixes most get_file errors by removing query argument (#33 @kuriwaki)
- Fix getting multiple files by id in
get_file()(#47 @adam3smith) - Temporary files created by
get_file()are automatically deleted.
- Export
initiate_sword_dataset(). (h/t Justin de Benedictis-Kessner) - Pass
key,server, and...arguments to internalget_dataverse()andget_dataset()calls. - Tests now run with an explicit empty API key.
- Fixed a bug in internal function
parse_dataset(), related to capitalization. (#17) - Vignette uses 'remotes' package in place of the archived 'ghit' package. (#24 @wibeasley)
- Updated config for Travis-CI, such as switch to xenial Ubuntu release, specify repo's org, and specify covr parameters. (#25 @wibeasley)
- Added an
update_dataset_file()function and improved associated documentation. (#10)
- Added a provisional
add_dataset_file()function. (#10) - Reorganized some code.
- Noted that user-related functions are not implemented (yet). (#1)
- Change vignette workflow so that vignettes are pre-built. (#1)
- Removed XML dependency, updating all code to xml2 instead.
- Removed urltools dependency.
- Finished the "Data Archiving" vignette. (#1)
- Fixed some bugs in
dataverse_search() get_file()now unzips its results when multiple files are requested and returns them as a raw vector.- Finished the "Data Retrieval" vignette. (#1)
- Document
dataverse_search()in a vignette. (#1)
- Update README.
- Update roxygen.
- Add
print.dataverse_file()method. (#12) - Added a
dataverse_id.character()method. (#12)
- Fixed a bug in
api_url()related to parsing of the Dataverse server URL that was affected by an API change in urltools. (h/t John Little)
- Initial commit