How do I write function to convert a character vector to minutes?-CodePudding

I would like to write a function to convert a character vector in the "time_spent" column to a numeric vector of minutes.

#Import libraries
library(tidyr)
library(dplyr)
library(lubridate)

#Create dataframe
df <- data.frame(country = c("Russia", "China", "USA"),
                 time_spent = c("0:10", "1:20", "2:36"))

#Convert "time_spent" from factor to character
df$time_spent <- lapply(df$time_spent, as.character)

I can do this with the following code, but I want to learn how to write it as a function.

df %>% 
  mutate(minutes = paste(time_spent, "00", sep = ":"),
         minutes = hms(minutes),
         minutes = hour(minutes)*60   minute(minutes))

I know I've butchered this function and its application, but I cannot figure out how to get it to work.

#Function to convert "time_spent" vector into minutes
convert_to_minutes <- function(x) {
  df$x = paste(x, "00", sep = ":") %>% 
  df$x = hms(x) %>% 
  df$x = hour(x)*60   minute(x)
  }

#Apply function
df %>% 
  convert_to_minutes(df$time_spent)

Related question 2

How do I use map to apply my convert_to_minutes function to multiple time_spent vectors as in the following dataframe?

df2 <- data.frame(country = c("Russia", "China", "USA"),
                 time_spent = c("0:10", "1:20", "2:36"),
                 time_spent2 = c("0:15", "0:12", "1:47"),
                 time_spent3 = c("0:25", "3:45", "0:18"))

CodePudding user response：

A function to convert to minutes, in base R only, could be the following.

convert_to_minutes <- function(x) {
  f <- function(x) as.vector(as.integer(x) %*% c(60, 1))
  y <- strsplit(as.character(x), ":")
  sapply(y, f)
}

y <- c("0:10", "1:20", "2:36")
convert_to_minutes(y)
#[1]  10  80 156

To convert a data.frame column, with dplyr function mutate.

library(dplyr)

df %>%
  mutate(minutes = convert_to_minutes(time_spent))
#  country time_spent minutes
#1  Russia       0:10      10
#2   China       1:20      80
#3     USA       2:36     156

To convert multiple columns, use mutate(across(.)).

df2 %>%
  mutate(across(starts_with("time"), convert_to_minutes, .names = "minutes_{col}"))
#  country time_spent time_spent2 time_spent3 minutes_time_spent minutes_time_spent2 minutes_time_spent3
#1  Russia       0:10        0:15        0:25                 10                  15                  25
#2   China       1:20        0:12        3:45                 80                  12                 225
#3     USA       2:36        1:47        0:18                156                 107                  18

CodePudding user response：

As you already use lubridate, you can make a more efficient use with it. Instead of using lubridate::hms it also supports lubridate:hm. Another handy function is lubridate::as.duration which allows you to format in any measures, which in your case is minutes. Then just convert the duration to numeric and you are done.

x <- c("0:10", "1:20", "2:36")
as.numeric(as.duration(hm(x)), "minutes")
# [1]  10  80 156

Full answer to convert df2

df2 <- data.frame(country = c("Russia", "China", "USA"),
                 time_spent = c("0:10", "1:20", "2:36"),
                 time_spent2 = c("0:15", "0:12", "1:47"),
                 time_spent3 = c("0:25", "3:45", "0:18"))

library(lubridate)
library(data.table)
setDT(df2)

cols <- 2:length(names(df2))

df2[, (cols) := lapply(.SD, function(x) as.numeric(as.duration(hm(x)), "minutes")), .SDcols = cols]

CodePudding user response：

There are few typos and syntax errors in your function. You can use -

library(lubridate)
library(dplyr)

convert_to_minutes <- function(x) {
  x = paste(x, "00", sep = ":") 
  x = lubridate::hms(x) 
  x = x@hour *60   x@minute
  x
}

convert_to_minutes(df$time_spent)
#[1]  10  80 156

To apply it for multiple columns you can use across in tidyverse (or lapply in base R).

df2 %>%
  mutate(across(contains('time_spent'), convert_to_minutes))

#  country time_spent time_spent2 time_spent3
#1  Russia         10          15          25
#2   China         80          12         225
#3     USA        156         107          18