Question

> X1 <- data.frame(A=c("a1","a2","a3","a4"))
> X1
   A
1 a1
2 a2
3 a3
4 a4

> y <- data.frame(A=c("b1", " ", " ", "b2"," ", "b3"), B=c("c1","c2","c3","c4","X1","c5"))
> y
   A  B
1 b1 c1
2    c2
3    c3
4 b2 c4
5    X1
6 b3 c5

I want to do three things in table y.
(1) replace X1 in column B of the table y with the entries (i.e. a1…a4) in the Table X1.
(2) fill all empty spaces in column A of table y.
(3) add new column C in table y and fill all with the number “1”

The new table y is expected to be as follows:

   A  B C
1 b1 c1 1
2 b1 c2 1
3 b1 c3 1
4 b2 c4 1
5 b2 a1 1
6 b2 a2 1
7 b2 a3 1
8 b2 a4 1
9 b3 c5 1

Could you mind to suggest effective way of doing this, I have a large number of X1 table, not just one. I have to match the names of these tables with column B of table y. I could do it one-by-one, but I am confident that there are smarter way of doing this. Pls kindly help.

Answer 1

To fill out the A column:

y$A<-y$A[y$A!=" "][cumsum(y$A!=" ")]

To add the C column (come on, you could’ve found this yourself!):

y$C=1 #Duh

To expand the B column:
If the order is not important:

x1part<-y[rep(which(y$b=="X1"), each=nrow(X1)),]
x1part$B<-X1$A
cbind(y[y$b!="X1",], x1part)

If the order is important:

newy<-y[rep(seq(nrow(y)), ifelse(y$b=="X1", nrow(X1), 1)),]
newy$B[newy$B=="X1"]<-X1$A

Note that this is untested, but the ideas are there.