反向日期时间排序 [R]

Question

我正在尝试按 date.time(POSIXct 对象)订购 DF，最新的 date.time 在前

即反过来

df1<-structure(list(date.time = structure(c(1368066412, 1365895151, 
1365969188, 1366105495, 1367433149, 1367604510, 1367614808, 1367619743, 
1368005216, 1368011698, 1366244224, 1366414926, 1367513240, 1367624274, 
1367640074), class = c("POSIXct", "POSIXt"), tzone = ""), station = c("M1", 
"F1", "F3", "F4", "F5", "L1", "L2", "L4", "L5", "L7", "F1", "F3", 
"F4", "L2", "L4"), code = c(10184, 10888, 10888, 10888, 10888, 
10888, 10888, 10888, 10888, 10888, 10889, 10889, 10889, 10889, 
10889)), .Names = c("date.time", "station", "code"), row.names = c(NA, 
15L), class = "data.frame")

初始排序:

df1[with(df1, order(code, date.time)), ]

这工作正常，但我想要的是将 date.time 部分的顺序倒过来，首先订购最新的 date.time 戳记。

失败的尝试:

df1[with(df1, order(code, -date.time)), ]

负值不适用于 POSIXct 对象

df1[with(df1, order(code, rev(date.time))), ]

以一种奇怪的方式而不是按预期方式对 DF 进行排序

任何帮助都会很棒!

Answer 1

也许是这样的:

df2 <- df1[order(df1$code, -as.numeric(df1$date.time)), ]
df2    

unclass，as.vector 还将 POSIXct 对象转换为数字。 as.integer 也是一个替代方案。