3

I want to find 5 digit numbers that have four 5s. I do the following:

data <- str_split(as.character(x = 10000:99999), pattern = "")

but when I try to find the number of lists that contain four 5s

sapply(data, function(x) c("5", "5", "5", "5") %in% data[[x]]) it gives me an error "Error in data[[x]] : no such index at level 1":

What am I doing wrong? any other more elegant solution?

daniell
  • 169
  • 1
  • 9
  • would something like this work ? `grep("5{4}", data, value=TRUE)` ... actually this only grabs 4 in a row. ` – user20650 Dec 03 '21 at 23:11
  • number_5s<-sapply(1:length(data), function(x) sum("5" == data[x][[1]])) sum(number_5s==4) – jwzinserl Dec 03 '21 at 23:20
  • at last found a regex that seem to work, from [here](https://stackoverflow.com/questions/57772766/regex-to-find-a-character-is-present-n-number-of-times) : `grep("^(?:[^5]*5[^5]*){4}$", data, value=TRUE)` – user20650 Dec 03 '21 at 23:49

5 Answers5

2
grep('^(?=.{5}$).*?(5)(.*?\\1){3}', 10000:99999, value = TRUE, perl=TRUE)

 [1] "15555" "25555" "35555" "45555" "50555" "51555" "52555" "53555"
 [9] "54555" "55055" "55155" "55255" "55355" "55455" "55505" "55515"
[17] "55525" "55535" "55545" "55550" "55551" "55552" "55553" "55554"
[25] "55555" "55556" "55557" "55558" "55559" "55565" "55575" "55585"
[33] "55595" "55655" "55755" "55855" "55955" "56555" "57555" "58555"
[41] "59555" "65555" "75555" "85555" "95555"

Regex Explanation:

Use a lookahead to determine whether the number contains five digits. ie ^(?=.{5}$). Then determine whethere there are 4 5's .*?(5)(.*?\\1){3}

Onyambu
  • 67,392
  • 3
  • 24
  • 53
2

Try this?

> x <- 10000:99999

> x[nchar(gsub("[^5]", "", x)) == 4 & nchar(x)==5]
 [1] 15555 25555 35555 45555 50555 51555 52555 53555 54555 55055 55155 55255
[13] 55355 55455 55505 55515 55525 55535 55545 55550 55551 55552 55553 55554
[25] 55556 55557 55558 55559 55565 55575 55585 55595 55655 55755 55855 55955
[37] 56555 57555 58555 59555 65555 75555 85555 95555

Or, you can generalize it like below

# number of digits
n <- 5 
# number of 5s
m <- 4
# output
(x <- 10^(n - 1):(10^n - 1))[nchar(gsub("[^5]", "", x)) == m]
ThomasIsCoding
  • 96,636
  • 9
  • 24
  • 81
1

I think the simplest is

5*9-1
# 44

This works because out of the five digits, exactly one is not five (leaving four fives remaining). For each non-five digit it can take any one of 9 different values (i.e. 0 to 9, except 5). Then we subtract 1 because we don't allow 05555, which would be a four digit number.

dww
  • 30,425
  • 5
  • 68
  • 111
0

This should get you started. It will extract all the 5's from an expression and then count them. Add na logic.

 library(stringr)
 y <- nchar(str_extract(data, "[5]+" ))  = 4
Jim
  • 191
  • 6
0

This would be one way:

library(dplyr)
library(tibble)
enframe(10000:99999) %>% 
    mutate(
        name = 10000:99999,
        value = as.character(value),
        value = str_split(value, '')
    ) %>% 
    unnest(value) %>% 
    filter(value == 5) %>% 
    count(name) %>% 
    filter(n == 4) %>%
    select(-n)
# A tibble: 44 x 1
    name
   <int>
 1 15555
 2 25555
 3 35555
 4 45555
 5 50555
 6 51555
 7 52555
 8 53555
 9 54555
10 55055
# ... with 34 more rows
Baraliuh
  • 2,009
  • 5
  • 11