"R" write/export dataframe preserving the class information

Question

Short Question

After formatting a dataframe in R with class information like "Date", "Factor", and etc. how to save the dataframe while preserving the class information. Consequently how to load the information back to a new R Project with preserving the above class information.

Long Question

I have created following dataframe with class information,

names <- c("Adrew", "Vikie", "Penny", "Samantha") 
date <- as.Date(c("1984-12-28", "1986-10-18", "1986-11-30", "1987-11-30"))
gender <- c("M", "M", "S", "S")

df <- data.frame(names, date, gender)

df$gender <- factor(df$gender,
                   levels = c("M", "S"), 
                   labels = c(1,2), 
                   order = FALSE)

and the classes are factor, date, factor

method 01

then write the dataframe in csv format and read it

write.csv(df, "df.csv")
dfcsv <- read.csv("df.csv")

and the daraframe classes are changed

method 02

then tried to write the file using table export method

write.table(df, "df.R")
dftable <- read.table("df.R")

again the data frame classes had been changed

Question

is there a proper method to write or save a data frame with preserving the class information like date, factor, etc information intact, and conversely how to load it back.

Answer 1

The best way to achieve this would be to save the data as an R data file using either save() or saveRDS():

# option 1
save(df, file="data.Rdata")
load("data.Rdata")

# option 2
saveRDS(df, file="data2.Rds")
df <- readRDS("data2.Rds")

Answer 2

Description

It is possible to write ASCII text representation of an R object with the use of dput, dump, dget and source. And with this it is possible to preserve the class information

Method

saving the file

to save the file use the dget and/or dump function, and to make sure that the file preserve the class information use the argument control = "all"

the dump function take a list of dataframes as input which can be listed as c("df", "df1", "df3") and in the other hand dput will only output a single dataframe to a file.

dput

dput(df, file = "dput.R", control = "all")

dump

dump(c("df", "df2", "df3"), file = "dump.R", control = "all")

reading the file

to read the file use the dget function for dput files and source function for dump as mentioned,

dput -> dget

df3 <- dget(file =  "dput.R")

dump -> source

source("dump.R")