Parsing single digit months

[OmarGonD]

I wanted to answer this question on StackOverFlow, with "anytime".

This are the dates the have, the format is: "single digit month, day, full year"

2/10/2016  
4/4/2016  
5/8/2016  
10/1/2016

However, anydate() only works on the first entry, not the rest. In the PDF at CRAN, we find:

Issues
The Boost Date_Time library cannot parse single digit months or days. So while ‘2016/09/02’
works (as expected), ‘2016/9/2’ will not. Other non-standard formats may also fail.
The is a known issue (discussed at length in issue tick 5) where Australian times are off by an hour.
This seems to affect only Windows, not Linux.

So, apparently this is a known bug/issue. Are there anywork arounds?

Or we should just use as.POSIXct(df$final_date, format = "%d/%m/%Y") ?

anydate("2/10/2016")
[1] "2016-02-10"
anydate("4/4/2016")
[1] NA
anydate("5/8/2016")
[1] NA
anydate("10/1/2016")
[1] NA

R version 3.3.1 (2016-06-21)
Platform: x86_64-w64-mingw32/x64 (64-bit)
Running under: Windows 7 x64 (build 7601) Service Pack 1

locale:
[1] LC_COLLATE=Spanish_Peru.1252  LC_CTYPE=Spanish_Peru.1252   
[3] LC_MONETARY=Spanish_Peru.1252 LC_NUMERIC=C                 
[5] LC_TIME=Spanish_Peru.1252    

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods  
[7] base     

other attached packages:
[1] anytime_0.0.4 dplyr_0.5.0  

loaded via a namespace (and not attached):
[1] lazyeval_0.2.0  magrittr_1.5    R6_2.1.3       
[4] assertthat_0.1  rsconnect_0.4.3 DBI_0.5        
[7] tools_3.3.1     tibble_1.2      Rcpp_0.12.7

[eddelbuettel]

It is a shortcoming in Boost Date_Time. It just doesn't do it:

R> anytime:::testFormat("%m/%d/%Y", "10/11/2016")
[1] "2016-10-11 CDT"
R> anytime:::testFormat("%m/%d/%Y", "2/3/2016")
[1] NA
R> 

Sadly, there is nothing we can do here. My wording is a little off on the help page but this is documented. So I'll close this, ok?

The "good news" that base R does parse it:

R> as.Date("2/3/2016")
[1] "2-03-20"
R> 

[OmarGonD]

Yes, sure. No problem.

[eddelbuettel]

Trust me, I find it annoying as hell too. The hope for anytime was to parse all sane formats. But arguably single digit is not one. You could write a regexp which transform a single digit month or day to two and then parse that...

[alexanu]

Thank you for explanation. I'm having the same problem, but with hours (it doesn't recognize e.g 9:15, but does 09:15).

[eddelbuettel]

Try setting the R parser which should do it:

R> anytime("2018-9-20 7:24:30", useR=TRUE)
[1] "2018-09-20 07:24:30 CDT"
R> 

It's too bad the Boost one doesn't by default but such is life.

[alexanu]

Thank you for quick reply. Unfortunately it didn't work out for me from the 1st attempt. I didn't want to dig why, so used lubridate::parse_date_time2("dmY HM") - was also pretty fast for 2mn rows.

[eddelbuettel]

lubridate, under the competition, also switched to a C++-based backend. Sadly by copying the library code I had already in the RcppCCTZ package. And out main difference is ease-of-use / absence of format requirement. If you bother with a format, you can just to strptime() in base R...

[alexanu]

:) thank you! you are doing really cool stuff. Unfortunately, I still do not understand all your packages, but maybe one day :)

[eddelbuettel]

Good news - this now works in master (and hence the next release probably at the end of the month):

 R> inp <- c("2/10/2016", "4/4/2016", "5/8/2016", "10/1/2016")
 R> library(anytime)
 R> anytime(inp)
 [1] "2016-02-10 CST" "2016-04-04 CDT" "2016-05-08 CDT"
 [4] "2016-10-01 CDT"
 R>

[grigory93]

Unfortunately, it won't still work for the strings containg single digit hour like this 9/27/2017 9:00 resulting in 2017-09-27 00:00:00

[eddelbuettel]

There are no claims in the documentation that incomplete or non-standard formats are supported.

The gold standard is still ISO 9601 so hell yeah to four digit years, two digit days, two digit hours, years-before-month-before-day, ... and so on. We rely on a standard parser from Boost (plus for good measure R's own if you want). If you have exotic or likely error-prone formats you may to do something custom.

[eddelbuettel]

The lack of seconds hurts as well. If you add them at least the R parser copes:

R> library(anytime)
R> anytime("9/27/2017 9:00:00 ", useR=TRUE)
[1] "2017-09-27 09:00:00 CDT"
R> anytime("9/27/2017 9:00:00 ")
[1] "2017-09-27 CDT"
R>