Geek Answers Handbook

How to change multiple date formats in one column

What I still have is a dataframe column with dates in different character formats. Several appear in the %d.%m.%Y pattern, some in %m/%d/%Y :

 data$initialDiagnose = as.character(data$initialDiagnose) data$initialDiagnose[1:10] [1] "14.01.2009" "9/22/2005" "4/21/2010" "28.01.2010" "09.01.2009" "3/28/2005" "04.01.2005" "04.01.2005" "9/17/2010" "03.01.2010"

I want them to be like Date () in the same format, but R refuses, of course.
So I first tried changing them with a separator:

 data$initialDiagnose[grep('/', data$initialDiagnose)] = as.character.Date(data$initialDiagnose[grep('/', data$initialDiagnose)], format = '%m/%d/%Y')

The analogue of "." dates. But that did not work.

How can I change them all into one format so that I can work with them?

+9
date format r
source share
3 answers
 a <- as.Date(data$initialDiagnose,format="%m/%d/%Y") # Produces NA when format is not "%m/%d/%Y" b <- as.Date(data$initialDiagnose,format="%d.%m.%Y") # Produces NA when format is not "%d.%m.%Y" a[is.na(a)] <- b[!is.na(b)] # Combine both while keeping their ranks data$initialDiagnose <- a # Put it back in your dataframe data$initialDiagnose [1] "2009-01-14" "2005-09-22" "2010-04-21" "2010-01-28" "2009-01-09" "2005-03-28" "2005-01-04" "2005-01-04" "2010-09-17" "2010-01-03"

In addition to this, the previous method is adapted to the situation when you have three (or more) different formats:

 data$initialDiagnose [1] 14.01.2009 9/22/2005 12 Mar 97 4/21/2010 28.01.2010 09.01.2009 3/28/2005 Levels: 09.01.2009 12 Mar 97 14.01.2009 28.01.2010 3/28/2005 4/21/2010 9/22/2005 multidate <- function(data, formats){ a<-list() for(i in 1:length(formats)){ a[[i]]<- as.Date(data,format=formats[i]) a[[1]][!is.na(a[[i]])]<-a[[i]][!is.na(a[[i]])] } a[[1]] } data$initialDiagnose <- multidate(data$initialDiagnose, c("%m/%d/%Y","%d.%m.%Y","%d %b %y")) data$initialDiagnose [1] "2009-01-14" "2005-09-22" "1997-03-12" "2010-04-21" "2010-01-28" "2009-01-09" "2005-03-28"
+17
source share

I like lubridate for ease of use:

 library(lubridate) # note added ugly formats below data <- data.frame(initialDiagnose = c("14.01.2009", "9/22/2005", "4/21/2010", "28.01.2010", "09.01.2009", "3/28/2005", "04.01.2005", "04.01.2005", "Created on 9/17/2010", "03 01 2010")) mdy <- mdy(data$initialDiagnose) dmy <- dmy(data$initialDiagnose) mdy[is.na(mdy)] <- dmy[is.na(mdy)] # some dates are ambiguous, here we give data$initialDiagnose <- mdy # mdy precedence over dmy data # initialDiagnose # 2009-01-14 # 2005-09-22 # 2010-04-21 # 2010-01-28 # 2009-09-01 # 2005-03-28 # 2005-04-01 # 2005-04-01 # 2010-09-17 # 2010-03-01
+15
source share

Since MattBagg's answer in 2012, lubridate added a parse_date_time function that is designed specifically for this situation and can solve this problem in one line:

 library(lubridate) data <- data.frame(initialDiagnose = c("14.01.2009", "9/22/2005", "4/21/2010", "28.01.2010", "09.01.2009", "3/28/2005", "04.01.2005", "04.01.2005", "Created on 9/17/2010", "03 01 2010")) parse_date_time(data$initialDiagnose, orders = c('mdy', 'dmy')) [1] "2009-01-14 UTC" "2005-09-22 UTC" "2010-04-21 UTC" "2010-01-28 UTC" "2009-01-09 UTC" [6] "2005-03-28 UTC" "2005-01-04 UTC" "2005-01-04 UTC" "2010-09-17 UTC" "2010-03-01 UTC"
0
source share

More articles:


All Articles