R Programming : Logical dataframe to actual dataframe - r

Question

I need to convert or manipulate the records based on the logical dataframe in R.
I want to match with original dataframe and populate only those values equal to true from original dataframe and null for false value and also maintain the dataframe structure as well. Please suggest
For eg :
Original dataframe
ID Name Title
1 John Mr
2 Mike Mr
3 Susan Dr
Logical Dataframe
ID Name Title
False False False
False True False
False False True
Expected Dataframe
ID Name Title
2 Mike <null>
3 <null> Dr

Answer 1

Here's a shot:
orig <- read.table(text="ID Name Title
1 John Mr
2 Mike Mr
3 Susan Dr", header = TRUE, stringsAsFactors = FALSE)
lgl <- read.table(text="ID Name Title
False False False
False True False
False False True", header = TRUE, stringsAsFactors = FALSE)
newdf <- mapply(function(d,l) { d[!l] <- NA; d; }, orig, lgl)
newdf
# ID Name Title
# [1,] NA NA NA
# [2,] NA "Mike" NA
# [3,] NA NA "Dr"
newdf[ rowSums(!is.na(newdf)) > 0, ]
# ID Name Title
# [1,] NA "Mike" NA
# [2,] NA NA "Dr"
Your expected output is inconsistent in that you have FALSE in your $ID column, but you keep them in your output. You can fix that by changing those to TRUE and changing the filter to rowSums(!is.na(newdf)) > 1.
Explanation:
mapply runs a function (named or anonymous) on one or more lists, like a "zipper" function. That is:
mapply(func, 1:3, 4:6, 7:9, SIMPLIFY=FALSE)
is equivalent to
list(func(1,4,7), func(2,5,8), func(3,6,9))
!is.na(newdf) creates a data.frame of the same dimensions/names, but all elements are logical.
since in general sum(<logical_vector>) returns a single integer of how many elements are true, rowSums(...) returns a vector, one element per row, where each element is the number of "trues" on that row.
... > 0 returns a logical vector, only passing the rows that have at least one non-NA element.
You said you wanted to always preserve $ID. In that case, you probably want to do (before process):
lgl$ID <- TRUE
and change the condition to ... > 1 to me "at least two non-NA elements, one of which we know is ID".