Most changes in this version are to update the package to work with the new ORCID API (v2.1
). (#37) (#40)
rorcid
now support OAuth authentication. We still recommend to not use OAuth, but to get a token and store that as an environment variable. See ?orcid_auth
for help (#26)orcid_activities()
, orcid_address()
, orcid_auth()
, orcid_bio()
, orcid_educations()
, orcid_email()
, orcid_employments()
, orcid_external_identifiers()
, orcid_fundings()
, orcid_keywords()
, orcid_other_names()
, orcid_peer_reviews()
, orcid_person()
, orcid_ping()
, orcid_researcher_urls()
, orcid_works()
data.table
for binding lists together into a data.frameidentifiers()
for new API. Includes better failure behavior on classes it doesn’t support (#34) (#39)orcid_id()
changed internally; now wraps the new function orcid_person()
(#41)httr
with crul
for HTTP requests. we have retained httr
only to do OAuth (#32)orcid_id()
loses its profile
parameter due to the ORCID API change. it does pass on parameters to orcid_person()
, so see that man fileworks()
now returns a tibble/data.frame instead of a list of itemsorcid_id()
function gains output for employment and funding (#24) (#29)is()
calls to inherits()
(#30)tibble
package now for compact data.frame outputs instead of internal code. an associated change in the output of both orcid()
and orcid_doi()
is that we now return a tibble (data.frame) instead of a data.frame as a slot in a list. we add how many results are returned from your search as an attribute on the data.frame. Access it like attr(out, "found")
(#25)http
to https
schemesummary.or_id()
is now defunct. see ?rorcid-defunct
httr >= v1.1.0
(#23)dplyr
tidy data.frame internal code (#21)httr::content()
to parse to text, then read JSON manually using jsonlite
& to always set encoding
explicitly in the same calls (#22)as.orcid()
and presumably other function calls by requiring httr >= v1.1.0
because older versions cause a problem when parsing responses (#23) thanks @ericwatt