You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I think this adds to the complication of the tests, but I thought it would be worthwhile to allow testing things like left_join, right_join, full_join, and other multi-dataset operations. In order to do this, we may need to up the dataset count to 2.
However, the more useful feature (and with the right dataset, maybe rendering multiple datasets irrelevant) would be to allow arbitrary expressions in the YAML test. Then you just need join-able columns in your test dataset
I.e. maybe something like the following in run_test:
if (verb=="verbatim") manip<-f# or some other rlang magic
Note that at present, we use pull at the end of test strings so that the equivalent test is only on a single column. If allowing full dataset comparison, we would likely need to evaluate whether equivalent is too liberal/strict.
I think this adds to the complication of the tests, but I thought it would be worthwhile to allow testing things like
left_join
,right_join
,full_join
, and other multi-dataset operations. In order to do this, we may need to up the dataset count to 2.However, the more useful feature (and with the right dataset, maybe rendering multiple datasets irrelevant) would be to allow arbitrary expressions in the YAML test. Then you just need join-able columns in your test dataset
I.e. maybe something like the following in
run_test
:And how it might be used...
The text was updated successfully, but these errors were encountered: