I have a dataframe like the following:
1 a
2 a
3 a
4 a
a a
x b
y b
b b
How can I use are to get something like:
a 1,2,3,4
b x,y
Consider that the identifier a and b are also listed in column 1 but should not be part of the concatenated string. Many thanks!
Based on the initial description,
aggregate(V1~V2, df1, toString)
# V2 V1
#1 a 1, 2, 3, 4
#2 b x, y
Suppose, if we need to remove the elements in first column ('V1') that are also present in the grouping column ('V2') before we paste the elements of V1 together, a couple of options are listed below:
library(data.table)
setDT(df2)[!V1 %chin% V2, toString(V1), by=V2]
# V2 V1
#1: a 1, 2, 3, 4
#2: b x, y
Or
library(dplyr)
df2 %>%
filter(!V1 %in% V2) %>%
group_by(V2) %>%
summarise(V1=toString(V1))
df1 <- structure(list(V1 = c("1", "2", "3", "4", "x", "y"),
V2 = c("a",
"a", "a", "a", "b", "b")), .Names = c("V1", "V2"),
class = "data.frame", row.names = c(NA, -6L))
df2 <- structure(list(V1 = c("1", "2", "3", "4", "a", "x", "y", "b"),
V2 = c("a", "a", "a", "a", "a", "b", "b", "b")), .Names = c("V1",
"V2"), class = "data.frame", row.names = c(NA, -8L))
%chin% was new.filter option and put that in the summarise. I was thinking this would be a bit more clear for the OP.