collate="rows" strips Date objects into numeric

[sdanielzafar]

library(tidyverse)
library(purrrlyr)

data("storms")

storms %>% 
    head(100) %>% 
    mutate(Date = as.Date(paste0(year, "-", month, "-", day))) %>% 
    by_row(function(.d) {select(.d, Date)}, .to = "new", .collate = "rows")

which yeilds:

# tibble [100 × 16]
   name   year month   day  hour   lat  long status              category  wind pressure ts_diameter hu_diameter Date        .row  Date
   <chr> <dbl> <dbl> <int> <dbl> <dbl> <dbl> <chr>               <ord>    <int>    <int>       <dbl>       <dbl> <date>     <int> <dbl>
 1 Amy    1975  6.00    27  0     27.5 -79.0 tropical depression -1          25     1013          NA          NA 1975-06-27     1  2003
 2 Amy    1975  6.00    27  6.00  28.5 -79.0 tropical depression -1          25     1013          NA          NA 1975-06-27     2  2003
 3 Amy    1975  6.00    27 12.0   29.5 -79.0 tropical depression -1          25     1013          NA          NA 1975-06-27     3  2003
 4 Amy    1975  6.00    27 18.0   30.5 -79.0 tropical depression -1          25     1013          NA          NA 1975-06-27     4  2003
 5 Amy    1975  6.00    28  0     31.5 -78.8 tropical depression -1          25     1012          NA          NA 1975-06-28     5  2004
 6 Amy    1975  6.00    28  6.00  32.4 -78.7 tropical depression -1          25     1012          NA          NA 1975-06-28     6  2004
 7 Amy    1975  6.00    28 12.0   33.3 -78.0 tropical depression -1          25     1011          NA          NA 1975-06-28     7  2004
 8 Amy    1975  6.00    28 18.0   34.0 -77.0 tropical depression -1          30     1006          NA          NA 1975-06-28     8  2004
 9 Amy    1975  6.00    29  0     34.4 -75.8 tropical storm      0           35     1004          NA          NA 1975-06-29     9  2005
10 Amy    1975  6.00    29  6.00  34.0 -74.8 tropical storm      0           40     1002          NA          NA 1975-06-29    10  2005
# ... with 90 more rows

but Date column should be of Date class.

[lionel-]

Duplicate of #3. To solve this in the general case, this would require new infrastructure that is being developed in the vctrs package. But note that while purrrlyr is still maintained, it is no longer under active development.

[lorenzwalthert]

In this case, should a life-cycle badge be added to this repo?

[lionel-]

There is a lifecycle badge in the README.

[lorenzwalthert]

Sorry, was not passionate enough to wait for the full page to be loaded -.- Thanks.