CODEWITHSUNDEEP
r
Kickball
Kickball

Reputation: 153

Grep include NA values in results

I am trying to use grep to filter my data but also include NAs in the results, which are currently being dropped because they do not match the grep expression.

platform        x86_64-w64-mingw32
version.string  R version 3.5.3 (2019-03-11)

Sample data

   value expected_result actual_result
1  10001            Pass          Pass             
2      0            Pass          Pass
3      6            Pass          Pass
4  20004            Pass          Pass
5     NA            Pass          Fail
6   4829            Fail          Fail
7    521            Fail          Fail
8     89            Fail          Fail
9  40012            Fail          Fail
10 47321            Fail          Fail

Code

df <- df[grep("(\\b\\d{1}\\b)|([0-9]{1}[0]{3}[0-9]{1})", df$value),]

Logic rules

1) The value will contain between 0 and 5 numeric characters.

2) The three values that should be retained are:

a) A single digit of data. (Example values 2 & 3)

b) No data or NA (Example values 5)

c) Five digits of data but the middle three digits must all be zeros. (Example values 1 & 4)

Upvotes: 3

Views: 1528

Answers (1)

akrun
akrun

Reputation: 887223

To include the NA rows, create a second condition with is.na and join them with | (OR) on the grepl

df[grepl("(\\b\\d{1}\\b)|([0-9]{1}[0]{3}[0-9]{1})", df$value)|is.na(df$value),]
#   value expected_result actual_result
#1 10001            Pass          Pass
#2     0            Pass          Pass
#3     6            Pass          Pass
#4 20004            Pass          Pass
#5    NA            Pass          Fail

Or make it a bit more compact

grepl("^\\d$|^([1-9]0{3}[1-9]$)", df$value)|is.na(df$value)

data

df <- structure(list(value = c(10001L, 0L, 6L, 20004L, NA, 4829L, 521L, 
89L, 40012L, 47321L), expected_result = c("Pass", "Pass", "Pass", 
"Pass", "Pass", "Fail", "Fail", "Fail", "Fail", "Fail"), actual_result = c("Pass", 
"Pass", "Pass", "Pass", "Fail", "Fail", "Fail", "Fail", "Fail", 
"Fail")), class = "data.frame", row.names = c("1", "2", "3", 
"4", "5", "6", "7", "8", "9", "10"))

Upvotes: 4

Related Questions