Changelog

metadata are updated if a file is outdated, but this step now requires the user’s approval
dim columns whose usage was previously deprecated are removed from idbank list dataframe (code breaks likely)
new idbanks from 2021 list are included in the internal package data
all files created in the checks are stored in a temporary directory
DESCRIPTION : LazyData is FALSE

if dataset names are provided, get clean metadata with get_idbank_list (code breaks likely). The metadata is then stored locally on the computer. The update is automatically done every 3 months, and can be manually triggered
use of dim columns in the idbank list is then deprecated
add metadata to raw data with the function add_insee_metadata
get columns’ name with the function get_column_title
search_insee function is now using the idbank and the dataset lists
option to choose the separator with the split_title function
vignette : deaths and births example added
backup idbank list stored inside the package
readsdmx dependency is used optionally but not by default
package now officially supported by INSEE, source code and issue reports are moved to INSEE’s official GitHub page

all data is cached, and then all queries are run once per R session
RapidXML C++ library is used by default in get_insee function through readsdmx package
get_insee_idbank splits automatically a big query into several queries of 400-idbank each
search_insee function : find datasets with a keyword
add_insee_title function : find INSEE title of an unlimited amount of idbanks
split_title function : split quickly the titles provided with the data
pkgdown website : https://hadrilec.github.io/insee/
code coverage with unit tests > 90%
rsdmx and tidyr dependencies dropped, readsdmx added

insee 1.1.7