0

Edit, previous question below: by changing long data format (from "%d %b %Y %H:%M:%S" to "%Y-%m-%d") certain months get lost in NAs (May, October and December).

Here is an reproducible example, where May and December become NA:

Time <- c("26 Apr 2019 12:14:05", "17 May 2019 12:07:31", "05 Jul 2019 07:43:51", "14 Dec 2019 21:48:47")

a <- c(1, 2, 3, 4)
b <- c(2, 4, 6, 1)
c <- c(3, 5, 9, 7)
d <- c(4, 3, 2, 1)

my.df <- data.frame(Time, A = a, B = b, C = c, D = d)

my.df$Time <- format(as.POSIXct(my.df$Time,"%d %b %Y %H:%M:%S",tz=""), format = "%Y-%m-%d")

my.df

        Time A B C D
1 2019-04-26 1 2 3 4
2       <NA> 2 4 5 3
3 2019-07-05 3 6 9 2
4       <NA> 4 1 7 1

Thanks for your help!

================== PREVIOUS QUESTION ==================

I am refering to an earlier question I already "solved": How can I simplify my date-time stamp in R?

So here is my data frame:

df <- read.table("Diviner 2019-12 (Pro26).csv", skip = 9, dec = ".", sep = ",", stringsAsFactors = F)
names(df) <- df[1,]
df <- df[-c(1),]
head(df)
                  Time       10cm       20cm       30cm       40cm      50cm      60cm      70cm      80cm
2 05 Apr 2019 09:46:13  20.706751  26.204191  23.663700  18.041510  3.507654  5.644918  3.947458  0.926415
3 11 Apr 2019 08:36:32  18.457157  25.762735  23.822021  18.596792  3.829793  6.639636  4.313009  1.002555
4 19 Apr 2019 09:24:16  17.224855  24.033939  21.703968  16.956991  3.507654  6.827912  4.417910  1.046471
5 26 Apr 2019 12:14:05  16.603245  22.160437  19.541498  15.735872  3.237127  6.169124  3.987147  1.002555
6 10 May 2019 07:40:20  19.685282  22.467394  19.208130  14.976073  3.184556  5.620616  3.888364  0.959796
7 17 May 2019 12:07:31  16.823887  23.139757  20.706751  16.868202  3.470846  5.500014  3.714201  0.985313

Data ranges from the beginning of April to the end of December. After converting I lose the months May, October and December without any obvious reason:

df$Time <- format(as.POSIXct(df$Time,"%d %b %Y %H:%M:%S",tz="UTC"), format = "%Y-%m-%d")
head(df)
        Time       10cm       20cm       30cm       40cm      50cm      60cm      70cm      80cm      90cm
2 2019-04-05  20.706751  26.204191  23.663700  18.041510  3.507654  5.644918  3.947458  0.926415  1.304021
3 2019-04-11  18.457157  25.762735  23.822021  18.596792  3.829793  6.639636  4.313009  1.002555  1.440603
4 2019-04-19  17.224855  24.033939  21.703968  16.956991  3.507654  6.827912  4.417910  1.046471  1.574125
5 2019-04-26  16.603245  22.160437  19.541498  15.735872  3.237127  6.169124  3.987147  1.002555  1.397690
6       <NA>  19.685282  22.467394  19.208130  14.976073  3.184556  5.620616  3.888364  0.959796  1.484311
7       <NA>  16.823887  23.139757  20.706751  16.868202  3.470846  5.500014  3.714201  0.985313  1.429803

Why does it work on all months but not on May, October and December?

Timo
  • 15
  • 6
  • 4
    Your code `format(as.POSIXct("17 May 2019 12:07:31","%d %b %Y %H:%M:%S", tz="UTC"), format="%Y-%m-%d")` works for me. Better provide `dput`, please read: https://stackoverflow.com/questions/5963269/how-to-make-a-great-r-reproducible-example/5963610#5963610 – jay.sf Jan 16 '20 at 15:08
  • Thanks, I created an reproducible example. – Timo Jan 16 '20 at 16:54
  • @Timo Where is the reproducible example? – Maurits Evers Jan 16 '20 at 20:49
  • @ jay.sf: Thanks jay.sf, in my case it does not work (and makes it even worse): Now the 17th of May is set to the current local date. @ Maurits: I have posted it as a comment and it was deleted. I have now edited the answer and put the example there. Sorry for the inconvenience, I just started on this board. – Timo Jan 17 '20 at 07:33

0 Answers0