Question

I’m trying to write a function that will load some xts data into R, and sort them by the date/time I specify. Maybe I’m trying a too simplistic method, but here is my code:

load.data <- function (x, time1, time2) #x is going to be the actual file name
{
    vector = get(load("C:/Users/username/Desktop/x"))
    sortvector = vector['time1/time2/']
    return (sortvector)
}

And when I execute it, I get the message:

In readChar(con, 5L, useBytes = TRUE) :
  cannot open compressed file 'C:/Users/username/Desktop/x', probable reason 'No such file or directory'

So, how can I make it so that my function will actually search for the file name, rather than actual generic ‘x’? I hope I was clear, and I would certainly appreciate any help greatly.

Answer 1

Use paste or paste0

load.data <- function (x, time1, time2) #x is going to be the actual file name
{
    vector = get(load(paste0("C:/Users/username/Desktop/", x)))
    sortvector = vector[paste(time1, time2, sep='/')]
    return(sortvector)
}

Also, look at ?FinancialInstrument:::saveSymbols.days and ?FinancialInstrument:::getSymbols.FI, or look at the code for those functions for examples of saving and loading xts objects

Edit

Here’s an example

set.seed(123)
dat <- xts(cumsum(rnorm(100)), as.POSIXct("2012-05-23") + 1:100*60*60*6)
tmpdir <- tempdir()
save(dat, file=file.path(tmpdir, "dat.RData"))
rm('dat') 

time1 <- "2012-05-24 10:00:00"
time2 <- "2012-05-25 11:00:00"
vector = get(load(paste(tmpdir, "dat.RData", sep="/")))
sortvector = vector[paste(time1, time2, sep='/')]
sortvector
#                         [,1]
#2012-05-24 12:00:00 -2.044404
#2012-05-24 18:00:00 -2.829308
#2012-05-25 00:00:00 -4.497250
#2012-05-25 06:00:00 -4.877477