NEWS

auk 0.7.1

added a helper function for processing bar chart data from eBird process_barcharts()

drop data.table dependency, no longer needed with readr speed improvements
fix bug arising from 'breeding bird atlas code' being renamed to 'breeding code' (issue #58)

get_ebird_taxonomy() now fails gracefully when eBird API is not accessible, fixing the CRAN check errors https://cran.r-project.org/web/checks/check_results_auk.html

Family common names now included in eBird taxonomy
auk_select() now requires certain columns to be kept
Better handling of file paths with prefix argument in auk_split()
Fixed bug causing undescribed species to be dropped by auk_rollup()
Add a ll_digits argument to filter_repeat_visits() to round lat/lng prior to identifying sites
Change of default parameters to filter_repeat_visits()
auk_bbox() now takes sf/raster spatial objects and grabs bbox from them

Dates can now wrap in auk_date(), e.g. use date = c("*-12-01", "*-01-31") for records from December or January
Fixed bug preventing dropping of age/sex column
Allow for a wider variety of protocols in auk_protocol()
Addresing some deprecated functions from rlang
Fixed bug causing auk_set_awk_path() to fail

Work around for bug in system2() in some R versions: https://bugs.r-project.org/bugzilla/show_bug.cgi?id=17508
Adding a filter for PROALAS checklists to auk_protocol()

rlang::UQ() and rlang::UQS() deprecated, switching to !! and !!!
auk_unique() now keeps track of all sampling event and observer IDs that comprise a group checklist

Updated to 2018 taxonomy; new function get_ebird_taxonomy() to get taxonomy via the eBird API
Better handling of taxonomy versions, many functions now take a taxonomy_version argument and use the eBird API to get the taxonomy
auk_getpath() renamed auk_get_awk_path(), and added auk_set_awk_path()
Added auk_set_ebd_path() and auk_get_ebd_path() to set and get the EBD_PATH environment variable. Now users only need to set this once and just refer to the file name, rather than specifying the full path every time.
Functions to prepare data for occupancy modeling: filter_repeat_visits() and format_unmarked_occu()
New auk_bcr() function to extract data from BCRs
Added bcr_codes data frame to look up BCR names and codes
"Area" protocol added to auk_protocol() filter.
auk_extent() renamed auk_bbox(); auk_extent() deprecated and redirects to auk_bbox()
auk_zerofill() now checks for complete checklists and gives option to not rollup
auk_rollup() now gives the option of keeping higher taxa via drop_higher argument
auk_clean() deprecated
Fixed package load error when EBD_PATH is invalid
Fixed bug when reading files with a blank column using readr

New function, auk_split(), splits EBD up into multiple files by species
New object, auk_sampling, and associated methods for working with the sampling data only
New function, auk_select(), for selecting a subset of columns
auk_date() now allows filtering date ranges across years using wildcards, e.g. date = c("*-05-01", "*-06-30") for observations from May and June of any year
New function, auk_state() for filtering by state
Now using AWK arrays to speed up country and species filtering; ~20% speed up when filtering on many species/countries
Allow selection of a subset of columns when filtering
Remove free text columns in auk_clean() to decrease file size
Updated to work with Feb 2018 version of EBD
Fixed broken dependency on countrycode package

eBird taxonomy update to August 2017 version, users should download the most recent EBD to ensure the taxonomy is in sync with the new package
Manually set AWK path with environment variable AWK_PATH in .Renviron file
auk_distance, auk_breeding, auk_protocol, and auk_project filters added
Users can now specify a subset of columns to return when calling auk_filter using the keep and drop arguments
Many changes suggested by rOpenSci package peer review process, see https://github.com/ropensci/onboarding/issues/136 for details
New vignette added to aid those wanting to contribute to package development