Reputation: 5392
My R script reads in CSV data of two columns, where the columns are timestamps and scalar values, with:
data = read.csv(dataFilePath, colClasses=c("charDate", "numeric"))
My timestamps are formatted e.g. "5/14/14 13:14". The class charDate
is defined to convert the timestamps:
setClass("charDate")
setAs("character", "charDate", function(from) strftime(from))
The datafiles I'm running are here. They all read in as expected, represented in the R dataframe data
as e.g. "2014-05-14 13:14:00", except for those in the realTweets/ directory. Why? For realTweets/, the timestamps in data
look like e.g. "2014-05-14", where the time info has not been preserved.
Upvotes: 0
Views: 127
Reputation: 5392
The issue was with strftime()
. As explained in this answer, strftime()
just wraps as.POSIXlt()
. Using as.POSIXlt()
directly fixed my problem:
setAs("character", "charDate", function(from) as.POSIXlt(from))
Upvotes: 1