match_name()
gains a new argument in its signature,sector_classification
, which is placed before the...
argument. Where users have used...
to pass additional unnamed arguments by position, they will need to update their code to pass these arguments by name so that they are not confused as an input tosector_classfication
. The new argumentsector_classification
is optional and defaults tor2dii.data::sector_classifications
. Explicitly passing asector_classification
tomatch_name()
, enables users to use their own sector classification systems to match loan books to the abcd, assuming thesector_classification
is mapped appropriately to PACTA sectors and has the same format asr2dii.data::sector_classifications
. Own sector classifications can no longer be passed tomatch_name()
via options.
- r2dii.match is now stable.
-
Complete deprecation of
ald
in favour ofabcd
(#399). -
match_name
gains argumentjoin_id
allowing an optional perfect join based on a mutual ID column betweenloanbook
andabcd
inputs, prior to attempting fuzzy matching (#135).
to_alias
can now handle strange encodings without error (#425, @kalashsinghal @Tilmon).
r2dii.match
has transferred to a new organization https://github.com/RMI-PACTA/.
- Maintenance release.
- New argument
abcd
ofmatch_name()
supersedes the argumentald
(#399).
- Maintenance release.
- Maintenance release.
-
With
options(r2dii.match.sector_classifications = own)
users can inject theirown
sector_classififications
instead of the defaultr2dii.data::sector_classifications
. With this feature, user may or may not choose to request theirsector_classifications
to be added to r2dii.data. This feature is experimental and may be dropped and/or become a new argument tomatch_name()
(#356 @georgeharris2deg @daisy-pacheco)`. -
match_name()
now errors if the columnid_loan
of the input loanbook has any duplicated valueid_loan
(@georgeharris2deg #349).
- Maintenance release.
match_name()
gains...
to pass additional arguments tostringdist::stringsim()
(@evgeniadimi #310).
prioritize()
with 0-row input now returns the input untouched (#284).- Fix
match_name()
: Remove dependency onnest_by()
from dplyr 1.0.0 (#303).
- Change license to MIT.
- Increment lifecycle badge to "Maturing".
- The website's home page now thanks funders.
- New article on using
match_name()
with large loanbooks. - The News tab of the website now shows all releases to date.
- New article "Calculating matching coverage" (#264).
match_name()
now outputs a new columnborderline
(#258).- New
crucial_lbk()
helps select the minimum loanbook columns formatch_name()
to run (#236). match_name()
now runs faster and uses less memory (@georgeharris2deg #214).match_name()
now convertsald$sector
to lower case before matching (@georgeharris2deg #257). It now returns identical output with, for example, either "POWER" or "power". Notice that the input "POWER" inald$sector
becomes "power" in the columnsector_ald
of the output.match_name()
now errors with a more informative message ifloanbook
has reserved columns --alias
,rowid
, orsector
(#233).
- Enforce dplyr >= 0.8.5 (#216).
- No longer import vctrs; it is unused.
This version includes only internal changes.
- First release on CRAN.