again something that would work only for eat because it assumes asymmetry.
eat(x,y, by = c(colx = toupper(coly))
it would require NSE of course, maybe it's misleading to use c, a temp column would be created for the join and deleted in the end. Using := we could use it on both sides and integrate in every join, however if we need to get rid of original cols we need to do it like this :
eat(x,y, -coly, by = c(colx = toupper(coly))
Taking into consideration conflict edge cases.
And make sure that the temp joining columns are computed before columns are removed
again something that would work only for
eatbecause it assumes asymmetry.eat(x,y, by = c(colx = toupper(coly))it would require NSE of course, maybe it's misleading to use
c, a temp column would be created for the join and deleted in the end. Using:=we could use it on both sides and integrate in every join, however if we need to get rid of original cols we need to do it like this :eat(x,y, -coly, by = c(colx = toupper(coly))Taking into consideration conflict edge cases. And make sure that the temp joining columns are computed before columns are removed