CODEWITHSUNDEEP
r
Javier
Javier

Reputation: 730

Manipulating Date with Conditional Statements

I'm trying to add and subtract days from my date-time but facing some issues.

Reproducible Example

df = data.frame(date1=c("2017-07-07", "2017-02-11", "2017-05-22")) 

library(lubridate)
df$date1 <- ymd(df$date1) + years(2)
df$day <- wday(df$date1, label=TRUE)

df$date1 <- as.Date(df$date1, "%Y-%m-%d")
df$day <- as.character(df$day)

Data

       date1 day 
1 2019-07-07 Sun   
2 2019-02-11 Mon       
3 2019-05-22 Wed

Problem

I'm trying to get my code to work in a manner that when the day is "Sunday", it will simply copy the same date. However, my code above returns "18084" instead of "2019-07-07".

Could someone explain to me where my issue is? 

library(dplyr)
df %>% mutate(newdate = ifelse(day == "Sun", as.Date(as.character(date1), "%Y-%m-%d"), 1))

       date1 day newdate
1 2019-07-07 Sun   18084
2 2019-02-11 Mon       1
3 2019-05-22 Wed       1

Upvotes: 2

Views: 42

Answers (1)

Ronak Shah
Ronak Shah

Reputation: 388982

The issue is you are trying to return objects of two different class in ifelse. If you check

str(df)
#'data.frame':  3 obs. of  2 variables:
# $ date1: Date, format: "2019-07-07" "2019-02-11" "2019-05-22"
# $ day  : chr  "Sun" "Mon" "Wed"

So date1 is of class "Date" and the else part in your ifelse contains 1 which is numeric. So now to bring both of them to the same class what R does is 

as.numeric(as.Date("2019-07-07"))
#[1] 18084

Dates are anyway represented internally as numbers. For example, 

as.Date(18084, origin = "1970-01-01") #would give
#[1] "2019-07-07"

What you probably need is

library(dplyr)
df %>% mutate(newdate = ifelse(day == "Sun", as.character(date1), "1"))

#       date1 day    newdate
#1 2019-07-07 Sun 2019-07-07
#2 2019-02-11 Mon          1
#3 2019-05-22 Wed          1

but this will change newdate to character class.

Upvotes: 2

Related Questions