Finding keys having value greater than 2

Question

• I have a key value pair as below which is in form of string

    > data$Frequency 1 A1:6,B1:4,AA1:3 2 BB1:2,AAA1:1 3 BBB1:1,A2:1,C2:3 4 D1:1 > class(data$Frequency) "character" FreqIndex<-which(colnames(data)=="Frequency") freq_string<-paste(as.character(unlist(data[FreqIndex])),collapse=",") print(freq_string) >"A1:6,B1:4,AA1:3,BB1:2,AAA1:1,BBB1:1,A2:1,C2:3,D1:1" 

• How can i find keys which have value greater than 2

   A1,B1,AA1,C2 

Answer 1

Here is another variation of strsplit where we split by : or , , then assuming that there are always key/value pairs, use the logical index recycling to subset the numeric values, check whether it is greater than 2 and subset the non-numeric elements

v1 <- unlist(strsplit(data$Frequency, "[:,]"))
v1[c(TRUE, FALSE)][as.numeric(v1[c(FALSE, TRUE)]) >2]
#[1] "A1"  "B1"  "AA1" "C2" 

---

Or we can use tidyverse options

library(tidyverse)
filtered <- data %>%
              separate_rows(Frequency, sep=",") %>%
              separate(Frequency, into = c('group', 'freq'), convert = TRUE) %>% 
              filter(freq>2) %>% 
              pull(group)
filtered
#[1] "A1"  "B1"  "AA1" "C2" 

condition <- paste(shQuote(filtered, type="cmd"), collapse=", ")
condition
#[1] "\"A1\" , \"B1\" , \"AA1\" ,\"C2\"" 

Answer 2

x <- "A1:6,B1:4,AA1:3,BB1:2,AAA1:1,BBB1:1,A2:1,C2:3,D1:1" #create the string
y <- sapply(strsplit(x, ","), strsplit, ":") #split the string into single keys and values

dat <- data.frame(key = rep(NA, 9), value = rep(NA, 9)) #prepare empty dataframe

## fill dataframe
for(i in 1:length(y)){
  dat[i, 1] <- y[[i]][1]
  dat[i, 2] <- y[[i]][2]
}

gives you

> dat
   key value
1   A1     6
2   B1     4
3  AA1     3
4  BB1     2
5 AAA1     1
6 BBB1     1
7   A2     1
8   C2     3
9   D1     1

and now dat$key[dat$value > 2] gives you

> dat$key[dat$value > 2]
[1] "A1"  "B1"  "AA1" "C2"