Replacing empty quotes with latest non-empty character

Question

The question is: how to fill empty quotes with previous row value in r? I have the following character array:

a=c("hello", "", "", "g_joy", "hello_w", "", "", "", "baby__", "rose", "samanthaberry11", 
    "eltonjames", "", "", "andrewger", "Ironman", "cec_sabry")

The desired result is, in this case:

>a
[1] "hello"           "hello"                "hello"                "g_joy"          
[5] "hello_w"         "hello_w"                "hello_w"                "hello_w"               
[9] "baby__"          "rose"            "samanthaberry11" "eltonjames"     
[13] "eltonjames"                "eltonjames"                "andrewger"       "Ironman"        
[17] "cec_sabry"

I was thinking of filling the vector using reduce:

xx = Reduce(function(x,y) if (y==' ') x else y, a)
b=cbind(xx,a)

Anyway, I did not obtain the desired result (the obtained result is in the first column):

> b
     xx          a                
[1,] "cec_sabry" "hello"          
[2,] "cec_sabry" ""               
[3,] "cec_sabry" ""               
[4,] "cec_sabry" "g_joy"          
[5,] "cec_sabry" "hello_w"        
[6,] "cec_sabry" ""               
[7,] "cec_sabry" ""               
[8,] "cec_sabry" ""               
[9,] "cec_sabry" "baby__"         
[10,] "cec_sabry" "rose"           
[11,] "cec_sabry" "samanthaberry11"
[12,] "cec_sabry" "eltonjames"     
[13,] "cec_sabry" ""               
[14,] "cec_sabry" ""               
[15,] "cec_sabry" "andrewger"      
[16,] "cec_sabry" "Ironman"        
[17,] "cec_sabry" "cec_sabry"

you're testing for `' '` not `''`: `Reduce(function(x,y) if (y=='') x else y, a, accumulate=TRUE)` — rawr, Apr 20 '21 at 07:22
semicolons are these -> ; Single quotes are these -> ' double quotes are these -> " — 1k monkeys and a single PC, Apr 20 '21 at 07:27

score 3 · Accepted Answer · answered Apr 20 '21 at 07:40

Try the base R code below

> Filter(nchar, a)[cumsum(!!nchar(a))]
 [1] "hello"           "hello"           "hello"           "g_joy"
 [5] "hello_w"         "hello_w"         "hello_w"         "hello_w"
 [9] "baby__"          "rose"            "samanthaberry11" "eltonjames"
[13] "eltonjames"      "eltonjames"      "andrewger"       "Ironman"
[17] "cec_sabry"

score 1 · Answer 2 · answered Apr 20 '21 at 07:32

library(zoo)
# Replace "" with <NA>
a[a == ""] <- NA
# Fill NA with last known value, keep leading NA's
na.locf(a, na.rm = FALSE)

# [1] "hello"           "hello"           "hello"           "g_joy"           "hello_w"         "hello_w"        
# [7] "hello_w"         "hello_w"         "baby__"          "rose"            "samanthaberry11" "eltonjames"     
# [13] "eltonjames"      "eltonjames"      "andrewger"       "Ironman"         "cec_sabry"

score 1 · Answer 3 · answered Apr 20 '21 at 07:36

This does the trick:


library(zoo)
`is.na<-`( a, a== "" ) %>% na.locf(na.rm=FALSE)

Using the is.na<- function like so is awkward, but it does offer a way to do this in one code line, which is nice for chaining.

It was designed to be used as demonstrated here.

Replacing empty quotes with latest non-empty character

3 Answers3