rdryad v1.0.0

BREAKING CHANGES

• Package redone to work with new Dryad API v2. Most functions are defunct, and there's three sets of new functions following the three major sets of API routes for datasets, versions, and files. See ?rdryad for more (#28) (#29)

rdryad v0.4.0

NEW FEATURES

• gains new function dryad_metadata() to download Dryad file metadata
• gains new function dryad_package_dois() to get file DOIs for a Dryad package DOI (a package can have many files) (#22)

MINOR IMPROVEMENTS

• dryad_files (formerly download_url()) now scrapes Dryad page to get URLs to Dryad files instead of using their API, which was not dependable (#26)
• dryad_fetch gains a parameter try_file_names (a boolean) which if TRUE we try to extract file names out of URLs (#26)

BUG FIXES

• fix to solr rdryad functions to hard code use of xml return format, and followlocation to follow any redirects (#27)

DEFUNCT

• download_url() is now defunct, see dryad_files()

NOTE

• two new pacakage dependencies: tibble and data.table

rdryad v0.3.0

NEW FEATURES

• Move to using solrium package instead of solr package
  for interaction with Dryad's Solr backend (#21) (#24)
• Now using crul instead of httr for HTTP requests (#23)
• gains two new functions handle2doi and doi2handle to
  convert between handles and DOIs, and DOIs and handles,
  respectively (#25)
• download_url function name has been changed to dryad_files, but
  you can still use download_url until the next version. In addition,
  download_url/dryad_files parameters id is changed to doi.

MINOR IMPROVEMENTS

• dryad_fetch is improved, and uses curl::curl_download instead of
  download.file. It now accepts >1 input URL, but destile length must
  equal number of urls.

rdryad 0.2.0

NEW FEATURES

• Re-worked most of the package.
• New package API, some methods are the same, but many are different. (#16)
• New functions (see functions starting with d_*()) to interact
  with Dryad Solr search engine (#10)
• OAI-PMH functions now using internally the oai package. (#14)

MINOR IMPROVEMENTS

• Slimmed down dependencies to a smaller set.
• Changed license from CC0 to MIT (#17)
• Added more tests (#18)
• Changed function to get files to only download them, and not attempt to
  read them into R, which introduces a very long dependency chain (#15)