Unnest doesn't work when nesting the index

[edo91]

I'm trying to perform the following nest and unnest actions.

library(tidyr)
library(tsibble)

data <- tsibbledata::hh_budget

data_nested <- data %>% nest(data = -Country)

data_nested %>% unnest(data)
#> Error in if (unknown_interval(interval) && (nrows > vec_size(key_data))) { : 
#>   missing value where TRUE/FALSE needed
#> In addition: Warning messages:
#> 1: In min(..., na.rm = TRUE) :
#>   no non-missing arguments to min; returning Inf
#> 2: In max(..., na.rm = TRUE) :
#>   no non-missing arguments to max; returning -Inf

I found out that the error comes from here:

vctrs::vec_unchop(data_nested$data)
#> Error in if (unknown_interval(interval) && (nrows > vec_size(key_data))) { : 
#>   missing value where TRUE/FALSE needed
#> In addition: Warning messages:
#> 1: In min(..., na.rm = TRUE) :
#>   no non-missing arguments to min; returning Inf
#> 2: In max(..., na.rm = TRUE) :
#>   no non-missing arguments to max; returning -Inf

I actually get the expected result if I do the following, however I can't define indices from unnest.

vctrs::vec_unchop(data_nested$data, indices = tsibble::key_rows(data))
#> # A tsibble: 88 x 7 [1Y]
#>     Year  Debt     DI Expenditure Savings Wealth Unemployment
#>    <dbl> <dbl>  <dbl>       <dbl>   <dbl>  <dbl>        <dbl>
#>  1  1995  95.7 3.72          3.40   5.24    315.         8.47
#>  2  1996  99.5 3.98          2.97   6.47    315.         8.51
#>  3  1997 108.  2.52          4.95   3.74    323.         8.36
#>  4  1998 115.  4.02          5.73   1.29    339.         7.68
#>  5  1999 121.  3.84          4.26   0.638   354.         6.87
#>  6  2000 126.  3.77          3.18   1.99    350.         6.29
#>  7  2001 132.  4.36          3.10   3.24    348.         6.74
#>  8  2002 149.  0.0218        4.03  -1.15    349.         6.37
#>  9  2003 159.  6.06          5.04  -0.413   360.         5.93
#> 10  2004 170.  5.53          4.54   0.657   379.         5.40
#> # ... with 78 more rows

I don't have the same error if Year (the index) is one of the nesting variables:

data %>% nest(data = -c(Year)) %>% unnest(data)
data %>% nest(data = -c(Country, Year)) %>% unnest(data)

However the outputs are tibbles instead of tsibbles.

---

Here sessionInfo():

R version 4.0.2 (2020-06-22)
Platform: x86_64-w64-mingw32/x64 (64-bit)
Running under: Windows 10 x64 (build 18363)

Matrix products: default

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

other attached packages:
[1] tsibble_0.9.2 tidyr_1.1.2  

loaded via a namespace (and not attached):
 [1] Rcpp_1.0.5        rstudioapi_0.11   knitr_1.30        magrittr_1.5      tidyselect_1.1.0  anytime_0.3.9    
 [7] R6_2.4.1          rlang_0.4.7       fansi_0.4.1       dplyr_1.0.2       tools_4.0.2       xfun_0.15        
[13] utf8_1.1.4        cli_2.0.2         tsibbledata_0.2.0 htmltools_0.5.0   ellipsis_0.3.1    assertthat_0.2.1 
[19] yaml_2.2.1        digest_0.6.25     tibble_3.0.3      lifecycle_0.2.0   crayon_1.3.4      purrr_0.3.4      
[25] vctrs_0.3.4       glue_1.4.2        evaluate_0.14     rmarkdown_2.3     compiler_4.0.2    pillar_1.4.6     
[31] generics_0.0.2    lubridate_1.7.9   pkgconfig_2.0.3  

[earowang]

Since {tidyr} v1.0.0, unnest() no longer works for tsibble and also makes it impossible to adapt for tsibble.

The data_nested is a tbl_df, and unnest.tbl_df() doesn't know how to properly handle a list of tsibbles. I can't provide unnest.tbl_df() in {tsibble} to overwrite it.

library(tidyr)
library(tsibble)
library(tsibbledata)

data <- hh_budget
data_nested <- data %>% nest(data = -Country)
class(data_nested)
#> [1] "tbl_df"     "tbl"        "data.frame"

^{Created on 2020-09-26 by the reprex package (v0.3.0)}

[Fuco1]

There was the function unnest_tsibble but is now marked as deprecated. What is the alternative?