R - Fill Column with values from any other columns

问题

I have a data frame of 5 columns: 4 columns have values and the 1 column that is empty. I want to fill the empty column with any value from any of the 4 columns.

Let's assume this is my dataframe df:

Col1 Col2 Col3 Col4 Col5
  11   11    
   2         2    2
       23
   4         4
       15        15

I want my result to look like this:

Col1 Col2 Col3 Col4 Col5
  11   11             11
   2         2    2    2
       23             23
   4         4         4
       15        15   15

EDIT I applied the answers provided by everyone, but it still isn't working for some reason. If it helps, this is the dput(head(df)) of my actual data:

structure(list(Treat_One = c("      ", "5 2012", "4 2008", "4 2010", 
"      ", "2 2008"), Treat_Two = c("8 2010", "5 2012", "4 2008", 
"4 2010", "8 2011", "2 2008"), Treat_Three = c("      ", "5 2012", 
"4 2008", "4 2010", "8 2011", "2 2008"), Treat_Four = c("      ", 
"      ", "      ", "      ", "      ", "      ")), .Names = c("Treat_One", 
"Treat_Two", "Treat_Three", "Treat_Four"), row.names = c(NA, 
6L), class = "data.frame")

Edit Included str(df)

'data.frame':   209 obs. of  4 variables:
 $ Treat_One  : chr  "      " "5 2012" "4 2008" "4 2010" ...
 $ Treat_Two  : chr  "8 2010" "5 2012" "4 2008" "4 2010" ...
 $ Treat_Three: chr  "      " "5 2012" "4 2008" "4 2010" ...
 $ Treat_Four : chr  "      " "      " "      " "      " ...

回答1:

Based on the new data provided by the OP we can remove leading/trailing white spaces with trimws

df$Treat_Four <- apply(df, 1, function(x) sample(x[trimws(x) != ""], 1))
df

#    Treat_One Treat_Two Treat_Three Treat_Four
#1              8 2010                 8 2010
#2    5 2012    5 2012      5 2012     5 2012
#3    4 2008    4 2008      4 2008     4 2008
#4    4 2010    4 2010      4 2010     4 2010
#5              8 2011      8 2011     8 2011
#6    2 2008    2 2008      2 2008     2 2008

Original Answer

We can use apply row-wise and take 1 sample of the element which is not equal to empty string

df$Col5 <- apply(df, 1, function(x) sample(x[x != ""], 1))
df
#  Col1 Col2 Col3 Col4 Col5
#1    1    1              1
#2    2         2    2    2
#3         3              3
#4    4         4         4
#5         5         5    5

If there are NA values and not blanks we can use the same logic

apply(df, 1, function(x) sample(x[!is.na(x)], 1))

回答2:

you can simply type as following:
df$Col5 <- 1:5
df$Col5 will create a Col5 in df and 1:5 just add the series number there.

回答3:

Try this:

df <- data.frame(col1 = c(1, NA, 3), col2 = c(1, 2, NA), col3 = c(NA, 2, 3),col4 = rep(NA, 3))
for (i in 1:nrow(df)) {
    df[i, 4] <- df[i, which(!is.na(df[i,]))][, 1]
}
df

That will produce:

> df <- data.frame(col1 = c(1, NA, 3), col2 = c(1, 2, NA), col3 = c(NA, 2, 3), col4 = rep(NA, 3))
> df
col1 col2 col3 col4
1    1    1   NA   NA
2   NA    2    2   NA
3    3   NA    3   NA
> for (i in 1:nrow(df)) {
+     df[i, 4] <- df[i, which(!is.na(df[i,]))][, 1]
+ }
+ df
+ 
col1 col2 col3 col4
1    1    1   NA    1
2   NA    2    2    2
3    3   NA    3    3

回答4:

Here is a vectorized option with max.col

df$Treat_Four <- df[1:3][cbind(1:nrow(df), max.col(sapply(df[1:3], trimws)!='', "first"))]
df
#   Treat_One Treat_Two Treat_Three Treat_Four
#1              8 2010                 8 2010
#2    5 2012    5 2012      5 2012     5 2012
#3    4 2008    4 2008      4 2008     4 2008
#4    4 2010    4 2010      4 2010     4 2010
#5              8 2011      8 2011     8 2011
#6    2 2008    2 2008      2 2008     2 2008

来源：https://stackoverflow.com/questions/42185450/r-fill-column-with-values-from-any-other-columns

标签

dataframe

multiple-columns