使用一个数字键和一个字符串键连接数据表

Question

我想通过两个关键列加入两个datatable：

df1=data.table(key1=c(1.2,1.1,1.9,4.6,2.3,3.6,0.9,2.8),key2=c("A","B","A","B","A","B","A","B"))
df2=data.table(key1=c(1,1,2,3),key2=c("A","B","B","B"),value=c("A","B","D","C"))

这是所需的输出：

   key1 key2 value
1:  0.9    A     A
2:  1.1    A     A
3:  1.2    B     B
4:  1.9    A    NA
5:  2.3    A    NA
6:  2.8    B     C
7:  3.6    B    NA
8:  4.6    B    NA

可以使用以下内容匹配具有最接近值的数字键：

# Join with closest value of Numeric Key, works fine

df1=data.table(key1=c(1.2,1.1,1.9,4.6,2.3,3.6,0.9,2,8),key2=c("B","B","A","A"))
    df2=data.table(key1=c(1,1,2,3),key2=c("A","B","B","B"),value=c("A","B","D","C"))
    setkey(df1,key1)
    setkey(df2,key1)
    df2[df1,roll="nearest"]

此外，如果数字键具有精确匹配，则问题可以再次解决：

df1=data.table(key1=c(1,1,2,5,2,4,1,3),key2=c("A","B","A","B","A","B","A","B"))
df2=data.table(key1=c(1,1,2,3),key2=c("A","B","B","B"),value=c("A","B","D","C"))
setkey(df1,key1,key2)
setkey(df2,key1,key2)
df2[df1]

解决方案不一定是data.table。

由于