Michael, this missing feature is definately a bug; AttributeList objects were meant to resemble data.frame objects but without having a row.names attribute (like the data.frame in R 2.4.0 will be); POSIXct vectors are allowed in data.frame's.
Did you actually solve it? Best regards, -- Edzer Michael Sumner wrote: >Hello, I've been unable to use "[[<-.AttributeList" to assign POSIXct vectors. > >It seems to be because >is.vector(Sys.time()) > >is FALSE. The test "is.vector(value)" is perhaps too strict ?? but I >don't see whether this is perhaps a problem with the datetimeClasses. > >Example follows. > >Cheers, Mike. > > >sessionInfo() >#Version 2.3.1 (2006-06-01) >#i386-pc-mingw32 > >#attached base packages: >#[1] "methods" "stats" "graphics" "grDevices" "utils" "datasets" >#[7] "base" > >#other attached packages: ># sp >#"0.8-16" > >al <- AttributeList(list(x = 1:10, y = gl(2, 5), t = Sys.time() + 1:10)) > > >al[["t"]] <- Sys.time() + 1:10 >#Error: is.vector(value) || is.factor(value) is not TRUE > >traceback() >#4: stop(paste(deparse(mc[[i + 1]]), " is not ", if (length(r) > ># 1) "all ", "TRUE", sep = ""), call. = FALSE) >#3: stopifnot(is.vector(value) || is.factor(value)) >#2: "[[<-.AttributeList"(`*tmp*`, "t", value = c(1153188132.868, ># 1153188133.868, 1153188134.868, 1153188135.868, 1153188136.868, ># 1153188137.868, 1153188138.868, 1153188139.868, 1153188140.868, ># 1153188141.868)) >#1: "[[<-"(`*tmp*`, "t", value = c(1153188132.868, 1153188133.868, ># 1153188134.868, 1153188135.868, 1153188136.868, 1153188137.868, ># 1153188138.868, 1153188139.868, 1153188140.868, 1153188141.868 ># )) > > >## the same problem occurs with >al[[3]] <- Sys.time() + 1:10 > >## and with > >al[["t"]] <- al[["t"]] > >## assigning numeric to numeric OK >al[["x"]] <- 2:11 > >## but assigning POSIX to same is not >al[["x"]] <- Sys.time() + 1:10 > > ># straight list version > >l <- list(x = 1:10, y = gl(2, 5), t = Sys.time() + 1:10) > >## works fine: >l[["t"]] <- l[["t"]] > >_______________________________________________ >R-sig-Geo mailing list >R-sig-Geo@stat.math.ethz.ch >https://stat.ethz.ch/mailman/listinfo/r-sig-geo > > _______________________________________________ R-sig-Geo mailing list R-sig-Geo@stat.math.ethz.ch https://stat.ethz.ch/mailman/listinfo/r-sig-geo