We can just follow their advice and cast mfargs to a list, as at a certain point we only ever evaluate elements in mfargs within data.
Please confirm this handles your warnings. I can't really test for this unless I figure out how to renew a session within a test, or create a test with a name that alphabetically comes first (and thus triggers the one warning per session warning).
Or I guess I could figure out if there's a way to always surface those warnings.
e: I just looked through how rlang tests these warnings and they ensure they are not on cran and then actually generate different deprecation warnings. I think we can rely on just checking warnings for now.
We can just follow their advice and cast
mfargsto a list, as at a certain point we only ever evaluate elements inmfargswithindata.Please confirm this handles your warnings. I can't really test for this unless I figure out how to renew a session within a test, or create a test with a name that alphabetically comes first (and thus triggers the one warning per session warning).
Or I guess I could figure out if there's a way to always surface those warnings.
e: I just looked through how
rlangtests these warnings and they ensure they are not on cran and then actually generate different deprecation warnings. I think we can rely on just checking warnings for now.