reshape::recast 不会使用 fill=NA 求和

Question

a <- data.frame(
variable=c("a","b","c"),
date=c("2015-01-01","2015-01-01","2015-01-02"),
value=c(1,2,3))

我明白

recast(a,date~variable,sum,fill=10000)
Using variable, date as id variables
>        date     a     b     c
>1 2015-01-01     1     2 10000
>2 2015-01-02 10000 10000     3

但是

recast(a,date~variable,sum,fill=NA)
Using variable, date as id variables

Error in vapply(indices, fun, .default) : values must be type 'logical', but FUN(X[[1]]) result is type 'double'

为什么会这样？

来自帮助

fill : value with which to fill in structural missings, defaults to value from applying fun.aggregate to 0 length vector

在这种情况下，当缺少值时，我希望 sum 为 NA。

Answer 1

指定您要使用的 NA 类型，它应该可以工作:

recast(a, date ~ variable, sum, fill = NA_real_)
# Using variable, date as id variables
#         date  a  b  c
# 1 2015-01-01  1  2 NA
# 2 2015-01-02 NA NA  3