How to replace NULL/? with 'None' or '0' in r - r

Question

DF1 is
ID CompareID Distance
1 256 0
1 834 0
1 946 0
2 629 0
2 735 1
2 108 1
Expected output should be DF2 as below (Condition for generating DF2 -> In DF1, For any ID if 'Distance'==1, put the corresponding 'CompareID' into 'SimilarID' column, for 'Distance'==0, ignore the corresponding 'CompareID')
ID SimilarID
1 None
2 735,108
Comparison is done correctly , but i got below output
ID SimilarID
1 ?
2 735,108
I understood that, as there are no 'CompareID' to put in 'SimilarID' - ? mark is displayed.
I want to replace this '?' with 'None' or '0'. Kindly help
In some cases, i observe that instead of '?' i can also see 'NULL' value.
Thanks !

Answer 1

Using the data.table package, where df is your original data ...
library(data.table)
setDT(df)[, .(SimilarID = if(all(Distance == 0)) "None"
else toString(CompareID[Distance == 1])), by = ID]
# ID SimilarID
# 1: 1 None
# 2: 2 735, 108
This follows your expected output by returning, by ID
"None" when all of the Distance column is zero
the CompareID values for when Distance is 1, as a comma-delimited string
Data:
df <- structure(list(ID = c(1L, 1L, 1L, 2L, 2L, 2L), CompareID = c(256L,
834L, 946L, 629L, 735L, 108L), Distance = c(0L, 0L, 0L, 0L, 1L,
1L)), .Names = c("ID", "CompareID", "Distance"), class = "data.frame", row.names = c(NA,
-6L))

Answer 2

Try the following using dplyr:
summarise.func <- function (Distance,CompareID) {
SimilarID <- CompareID[Distance == 1]
if (length(SimilarID)==0) "None" else paste0(SimilarID, collapse=",")
}
library(dplyr)
df2 <- df1 %>% group_by(ID) %>%
summarise(SimilarID=summarise.func(Distance,CompareID))
First, define a summarizing function summarise.func that:
Extracts the CompareID to a SimilarID vector if the Distance == 1.
If this SimilarID vector has elements, then return a string that are these CompareIDs collapsed with ","; otherwise return "None".
Then, use this summarise.func to summarise SimilarID grouped by ID.
Using your data:
print(df2)
### A tibble: 2 x 2
## ID SimilarID
## <int> <chr>
##1 1 None
##2 2 735,108

Answer 3

Using aggregate in base R:
df2 <- aggregate((CompareID*Distance)~ID, df, FUN=function(x)
ifelse(sum(x)>0, paste(x[x>0], collapse = ","), "None"))
names(df2) <- c("ID", "SimilarID") #if necessary
# ID SimilarID
#1 1 None
#2 2 735,108
CompareID*Distance ensures that CompareID will be ignored if Distance==0. Further, its grouped by ID, if sum is greater than 0, the non-zero values (x[x>0]) are comma-delimited and None, otherwise.