resource_metadata()now returns the actual metadata of a resource rather than a data.frame detailing the names and types of all fields available for the chosen resource. A new function provides the old behavior,
stringsAsFactors == TRUEdefault. This change aligns behavior of the package across compatible versions of R.
nrow_resource()returns the number of rows a resource contains.
all_packages()now makes better use of the underlying CKAN API and returns the package id in addition to the name. However, the function now returns a data.frame rather than a character vector. A default limit to the return of 1000 rows is also implemented, but can be set to a custom value by the user.
all_resources() now makes better use of the
underlying CKAN API and is significantly faster. The function also now
provides a warning and returns an empty data.frame if the arguments
specified result in no results being returned, replacing the previous
behavior which produced an error.
get_resource() now allows accepts character vectors
as arguments, allowing for the download of resources from multiple
packages. It also now allows the user to specify resource names as well
as ids. Multiple performance enhancements too.
get_data() now supports pagination. The
page_size argument allows users to specify the maximum
number of records to be returned per query. The default behavior is to
attempt to return all rows with a single query. SQL queries are now used
less often, improving performance.
get_data() are now by default ordered as
per the database primary key.
NEWS.mdfile to track changes to the package.