[R] logical test not functioning correctly on zoo series...what the hell?

[Gabor Grothendieck]

On Wed, Mar 14, 2012 at 12:43 PM, knavero <knavero at gmail.com> wrote:
> attached http://r.789695.n4.nabble.com/file/n4472408/dataout_2471_843.csv
> dataout_2471_843.csv
>


Here is how the problem can be presented in a self contained, minimal,
reproducible fashion (as per last two lines on every message to
r-help):

library(zoo)
Lines <- "Meter ID           ,Date / Time           ,KW(ch: 1  set:0)

MCAS-B-2471        ,2/11/12 12:45         ,0.08
MCAS-B-2471        ,2/11/12 13:00         ,8.92
MCAS-B-2471        ,2/11/12 13:15         ,7.2           "

# poster's read.zoo statement except for use of text=Lines
rawData <- read.zoo(text = Lines,  header = TRUE,
       FUN = as.chron, format = "%m/%d/%Y %H:%M", index.column = 2,
       sep = ",", aggregate = function(x) tail(x, 1))

>From this we see that its trying to read character data into a zoo
object which is not allowed.  We can either do this to remove the
offending column (note that having dropped the character column the
index column becomes column 1 so we can drop the index= argument):

tail1 <- function(x) tail(x, 1)
fmt <- "%m/%d/%Y %H:%M"

rawData <- read.zoo(text = Lines, FUN = as.chron, format = fmt,
	header = TRUE, sep = ",", aggregate = tail1,
	colClasses = c("NULL", NA, NA))

or if we want to split it into columns with one column per Meter ID we
could do it like this (which in this case will give the same result as
there is only one Meter ID in Lines):

rawData <- read.zoo(text = Lines, FUN = as.chron, format = fmt,
	split = 1, index = 2, header = TRUE, sep = ",", aggregate = tail1)


-- 
Statistics & Software Consulting
GKX Group, GKX Associates Inc.
tel: 1-877-GKX-GROUP
email: ggrothendieck at gmail.com