R › R help

Can I specify POSIX[cl]t column classes inside read.csv?

Classic

List

Threaded

3 messages Options

Thomas Levine-2

Can I specify POSIX[cl]t column classes inside read.csv?

I'm loading a nicely formatted csv file.

#!/usr/bin/env Rscript
kpi <- read.csv(
# This is a dump of the username, date_joined and last_login columns
# from the auth_user Django table.
'data/2012-04-23.csv',
colClasses = c('character')
)
print(kpi[sample(nrow(kpi), 3),2:3])

Here's what the three rows I printed look like.

last_login date_joined
2012-02-22 02:44:11 2011-09-19 03:07:35
2011-09-16 01:34:41 2011-09-16 01:34:41
2011-07-02 20:29:17 2011-07-02 20:29:17

Once I load them, I'm converting the datetimes to datetimes.

kpi$last_login <- as.POSIXlt(kpi$last_login)
kpi$date_joined <- as.POSIXlt(kpi$date_joined)

Can I do this inside of read.csv by specifying colClasses? It's
obviously not a problem if I can't; it just seems like I should be
able to.

Note that the following doesn't work because it doesn't save the times.

colClasses = c('character', 'Date', 'Date')

Thanks

Tom

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

David Winsemius

Re: Can I specify POSIX[cl]t column classes inside read.csv?

On Apr 23, 2012, at 11:48 AM, Thomas Levine wrote:

> I'm loading a nicely formatted csv file.
>
> #!/usr/bin/env Rscript
> kpi <- read.csv(
> # This is a dump of the username, date_joined and last_login
> columns
> # from the auth_user Django table.
> 'data/2012-04-23.csv',
> colClasses = c('character')
> )
> print(kpi[sample(nrow(kpi), 3),2:3])
>
> Here's what the three rows I printed look like.
>
> last_login date_joined
> 2012-02-22 02:44:11 2011-09-19 03:07:35
> 2011-09-16 01:34:41 2011-09-16 01:34:41
> 2011-07-02 20:29:17 2011-07-02 20:29:17
>
> Once I load them, I'm converting the datetimes to datetimes.
>
> kpi$last_login <- as.POSIXlt(kpi$last_login)
> kpi$date_joined <- as.POSIXlt(kpi$date_joined)
>
> Can I do this inside of read.csv by specifying colClasses?

Possibly. If there is an "as" function for a particular class, it can
be used in the colClasses vector of read.* functions. It appears that
your input file might have the right combination of formats and
separators for this to succeed.

> It's
> obviously not a problem if I can't; it just seems like I should be
> able to.
>
> Note that the following doesn't work because it doesn't save the
> times.
>
> colClasses = c('character', 'Date', 'Date')

Try instead:

colClasses = c('character', 'POSIXlt', 'POSIXlt')

--

David Winsemius, MD
West Hartford, CT

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

Jeff Newmiller

Re: Can I specify POSIX[cl]t column classes inside read.csv?

I recommend not putting POSIXlt vectors in data frames because of memory use and added complexity of the resulting data frame. That is, use

colClasses = c('character', 'POSIXct', 'POSIXct')

instead. The POSIXlt values will still be created as temporary variables for reading in, but the data frame will store only the simpler and more compact type for later use.
---------------------------------------------------------------------------
Jeff Newmiller The ..... ..... Go Live...
DCN:<[hidden email]> Basics: ##.#. ##.#. Live Go...
Live: OO#.. Dead: OO#.. Playing
Research Engineer (Solar/Batteries O.O#. #.O#. with
/Software/Embedded Controllers) .OO#. .OO#. rocks...1k
---------------------------------------------------------------------------
Sent from my phone. Please excuse my brevity.

David Winsemius <[hidden email]> wrote:

>
>On Apr 23, 2012, at 11:48 AM, Thomas Levine wrote:
>
>> I'm loading a nicely formatted csv file.
>>
>> #!/usr/bin/env Rscript
>> kpi <- read.csv(
>> # This is a dump of the username, date_joined and last_login
>> columns
>> # from the auth_user Django table.
>> 'data/2012-04-23.csv',
>> colClasses = c('character')
>> )
>> print(kpi[sample(nrow(kpi), 3),2:3])
>>
>> Here's what the three rows I printed look like.
>>
>> last_login date_joined
>> 2012-02-22 02:44:11 2011-09-19 03:07:35
>> 2011-09-16 01:34:41 2011-09-16 01:34:41
>> 2011-07-02 20:29:17 2011-07-02 20:29:17
>>
>> Once I load them, I'm converting the datetimes to datetimes.
>>
>> kpi$last_login <- as.POSIXlt(kpi$last_login)
>> kpi$date_joined <- as.POSIXlt(kpi$date_joined)
>>
>> Can I do this inside of read.csv by specifying colClasses?
>
>
>Possibly. If there is an "as" function for a particular class, it can
>be used in the colClasses vector of read.* functions. It appears that
>your input file might have the right combination of formats and
>separators for this to succeed.
>
>
>
>> It's
>> obviously not a problem if I can't; it just seems like I should be
>> able to.
>>
>> Note that the following doesn't work because it doesn't save the
>> times.
>>
>> colClasses = c('character', 'Date', 'Date')
>
>
>Try instead:
>
>colClasses = c('character', 'POSIXlt', 'POSIXlt')
>
>
>--
>
>David Winsemius, MD
>West Hartford, CT
>
>______________________________________________
>[hidden email] mailing list
>https://stat.ethz.ch/mailman/listinfo/r-help
>PLEASE do read the posting guide
>http://www.R-project.org/posting-guide.html
>and provide commented, minimal, self-contained, reproducible code.

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.