Closed renkun-ken closed 10 years ago
The current implementation forces to evaluate the lhs value and thus does not support first-argument piping with lhs unevaluated.
> library(pipeR) > (mpg > mean(mpg)) %>>% subset(x = mtcars) Error in eval(as.call(c(fun[1L], quote(.), fun[-1L])), envir = list(. = x), : object 'mpg' not found > (mpg) %>>% with(data = mtcars) Error in eval(as.call(c(fun[1L], quote(.), fun[-1L])), envir = list(. = x), : object 'mpg' not found > mpg %>>% (with(mtcars,.)) Error in eval(expr, list(. = .), envir) : object 'mpg' not found > (mpg > mean(mpg)) %>>% (subset(mtcars,.)) Error in eval(expr, list(. = .), envir) : object 'mpg' not found
The current implementation forces to evaluate the lhs value and thus does not support first-argument piping with lhs unevaluated.