data.table lappy convert date from numeric yyyymmdd to date format

Advertisements

Given the following DT:

Date1 <- c(20210103, 20220805, 20191231)                       
Amount <- c(10,20,30)                     
Date2 <- c(20190101, 20180220, 20240930) 
dt_1 <- data.table(Date1, Amount, Date2) 
dt_1

I am trying to convert the numeric yyymmdd to date format:

date_cols <- c(Date1, Date2)
dt_1 <- dt_1[, (date_cols) := lapply(.SD, as.Date), .SDcols = date_cols]

can you help me to fix it?

>Solution :

  1. date_cols should be defined as c("Date1", "Date2")
  2. Since date1 and date2 are numbers, change them to character first before converting them to date. Additionally, you need to supply the format the dates are present in your data.
library(data.table)

date_cols <- c("Date1", "Date2")

dt_1[, (date_cols) := lapply(.SD, \(x) as.Date(as.character(x), '%Y%m%d')), 
       .SDcols = date_cols]

#        Date1 Amount      Date2
#1: 2021-01-03     10 2019-01-01
#2: 2022-08-05     20 2018-02-20
#3: 2019-12-31     30 2024-09-30

Or you may also use lubridate::ymd which works without changing dates to characters and supplying the format explicitly.

dt_1[, (date_cols) := lapply(.SD, lubridate::ymd), .SDcols = date_cols]

Leave a ReplyCancel reply