Closed briatte closed 5 years ago
Hi @expersso
I'm reviewing my code from last month. Anything you need me to add to this PR?
Hi @briatte,
Sorry, I received this during a very busy period, so I must have missed it.
The changes look great. Should we perhaps add some details in the documentation for the relevant functions that they now accept proper DOIs as well as Gesis identifiers?
Hi @expersso
Don't be sorry, I'm also knee-deep into (mostly late) work.
I can update the docs, sure -- just leave a comment here to remind me to do so (I'm travelling for conferences right now) :-)
Alright, forget my previous comment, I updated the documentation for the doi
argument in the functions where it appears -- see above!
This PR does two things:
doi
is an actual DOI instead of a GESIS identifier, it converts it to a GESIS identifier.The PR adds a new internal function,
get_gesis_id
, to do so. It useshttr
, which is already imported by the package, and is otherwise coded in base R in order to avoid adding dependencies.I refrained from adding any additional checks on
doi
, although GESIS identifiers are, it seems, always made of four digits, so that could also get checked at some point.I also refrained from changing the rest of the codes, including the messages, that call GESIS identifiers "DOI", in order not to disrupt the discussion in #12 about this.