将行添加到data.frame中，但仅添加到某些列中

Question

I wish to create a data.frame and add rows but ONLY to certain columns. I want the remaining columns to be specified as NAs.

What I have is as follows:

#create empty data frame
x=data.frame(matrix(nrow=0,ncol=3))
> x
[1] X1 X2 X3
<0 rows> (or 0-length row.names)

#The data and its corresponding columns
> y=data.frame(col=c("X1","X3"),val=c(20,30))
> y
  col val
1  X1  20
2  X3  30
#select those columns (works fine)
> x[,y$col]
[1] X1 X2
<0 rows> (or 0-length row.names)
#select and create a row for those columns and corresponding values
> x[,y$col]=y$val
Error in Summary.factor(1:2, na.rm = FALSE) : 
  min not meaningful for factors

So the question is how would you (efficiently) create the new rows. Im guessing there ought to be a way to do it without specifying the NAs yourself.

Answer 1

One of the problems is that your col variable in y is a factor. This can easily be undone:

y=data.frame(col = c("X1","X3"), val = c(20,30), stringsAsFactors=FALSE)

Next, when you do

x[,y$col]=y$val

You don't specify a row number. Doing this would work though:

x[1,y$col] <- y$val

x

#   X1 X2 X3
# 1 20 NA 30

But then you'll soon run into problems if you have other X1's and X2's to write to the x dataframe. So your algorithm might need some brush-up, unless this is really what you need.