- Bug fix - read_erp() no longer returns a grouped data frame. Thanks to @nerskin for the fix.
- New read_erp() function added thanks to Francis Markham. Loads estimated resident population data.
- The experimental
readabs::read_api()
function no longer coerces columns describing the data categories to numeric. Thanks to @kletts. - read_lfs_datacube() now able to fetch modelled SA4 labour force estimates, thanks to @AngusMoore
- Bug fix to read_job_mobility()
- Fix to reflect changed ABS API URL
- read_payrolls() updated to work with reconfigured (less frequent) ABS payrolls data.
- Documentation clarified to note that
release_date
arg to read_abs() does not reliably work for annual data. - Refresh internal data
- An error inadvertently introduced in 0.4.15 is fixed. This error would only affect a small minority of users.
- read_lfs_datacube() convenience function added
- New environment variable "R_READABS_DL_METHOD" can be set. When set, this is passed to the
method
argument ofdownload.file()
. Useful on networks where a method such as "wininet" must be used.
- Fixes made to read_payrolls() to reflect changes by the ABS
- read_api() and related experimental functions added by @kintob (thank you!) to work with data from the ABS.Stat API
- Documentation expanded and improved
- The ability to read sub-state estimates of payroll jobs using
read_payrolls()
has been removed, in light of the ABS's removal of this data from the release.
- Added read_job_mobility()
- Added read_abs_url()
- Added a
release_date
argument toread_abs()
- Bug fixes
- Bug fixes
- New options available in read_payrolls(). Thanks to @martintburgess.
- Fixes to adapt to changes the ABS made to its time series spreadsheets without warning. Thanks to @Henry-DJPR.
- Fix to ensure read_abs_local() works with .xlsx files as well as .xlsx. Thanks to @hamgamb.
- "industry_wages" is no longer an option in read_payrolls() as it has been removed from the data
- Bug fixes
read_abs_sdmx()
function hard deprecated.check_latest_date()
function to check latest observation date for ABS time seriesread_lfs_grossflows()
convenience function addedutils::download.file()
used rather thancurl::curl_download()
- Residual upstream dependence on curl eliminated
- Workaround implemented for changes to ABS Weekly Payrolls release
search_catalogues()
function added to make data cubes easier to find
- Internal changes to the way files are downloaded
- Improvements to the
read_awe()
convenience function (able to get more data)
- Package dependencies reduced
- Bug fixed in
read_payrolls()
- New
read_payrolls()
convenience function added - New
read_awe()
convenience function added - Package logo added
- Bug fixes, including addressing a case where an ABS URL has a space in it
- minor fixes to unit tests
- download_abs_data_cube() and associated functions revamped for new ABS website
- Speed improvements
- get_abs() removed
- Minor internal change to prepare readabs for the new ABS website
- New download_abs_data_cube() function added courtesy of David Diviny
- Error when using dev version of vctrs package rectified
- Alternative method now used to check for internet access when running
read_abs()
; the function now works in a broader range of IT environments - README and vignette refined
- Default path for
read_abs()
is now defined by an environmental variable; thanks to Hugh Parsonage for the idea and implementation.
separate_series()
function gains adrop_nas
argument, thanks to Sam Gow for the suggestion.- Added more unit tests and reorganised tests
- New
separate_series()
function thanks to David Diviny, which splits theseries
column of tidied ABS time series into multiple components - New
read_cpi()
convenience function to get the CPI index numbers - Files read with
read_abs()
are now stored in a subdirectory ofpath
corresponding to the catalogue number - New
series_id
argument toread_abs()
allows users to get specific time series using their unique identifiers - Order of arguments to
read_abs()
have changed slightly, with newseries_id
argument added - Order of arguments to
read_abs_local()
have changed, newcat_no
argument added,filenames
argument works as before, but the argument order has changed - Fixed file path error when using
read_abs(retain_files = FALSE)
on Windows get_abs()
now deprecated; useread_abs()
instead (it has identical functionality)
read_abs()
now checks for internet access and returns a comprehensible error if not present- New function
read_abs_seriesid()
gets data corresponding to unique ABS time series IDs - Thanks to @HughParsonage for fixing the vignette index and suggesting other fixes
- ABS API was (silently!) updated to use https rather than http; readabs now works with this
- added a retain_files option (default = TRUE) to read_abs()
- Merged with
getabs
package - New core function (
read_abs()
) to download, import and tidy data in one step - New function (
read_abs_local()
) to import and tidy locally-stored spreadsheets read_abs_data()
is now soft-deprecated and will be removed in a future release
- Matt Cowgill is the new maintainer and author of
readabs
- Fixed issue (#1) with blank column names that arose from new name repair behaviour in tibble 2.0.0
- Add descriptive information to
read_abs_sdmx()
- Delete
read_abs_codebook()
- Create vignette and website using pkgdown
- Update documentation and available data
- Name change from abs to readabs