write_dta can't save labelled class without labels

[cimentadaj]

According to the documentation of labelled, the argument labels accepts a NULL, which is reasonable, given that you might want to save a column with no labels but an actual label. write_dta throws a weird error when saving with the labels argument to NULL.

library(haven)

# Note this this throws a warning but according to the documentation of `labelled`,
# it accepts a NULL
x <- labelled(1:20, labels = NULL, label = 'This is clearly a integer')
#> Warning in is.na(object): is.na() applied to non-(list or vector) of type
#> 'NULL'

x
#> <Labelled integer>: This is clearly a integer
#>  [1]  1  2  3  4  5  6  7  8  9 10 11 12 13 14 15 16 17 18 19 20

write_dta(data.frame(x = x), tempfile(fileext = '.dta'), version = 14)
#> Error in write_dta_(data, normalizePath(path, mustWork = FALSE), version = stata_file_format(version)): Not compatible with requested type: [type=NULL; target=integer].

# Adds toy labels
x <- labelled(1:20, labels = c('M' = 1), label = 'This is clearly a integer')

write_dta(data.frame(x = x), tempfile(fileext = '.dta'), version = 14)

devtools::session_info()
#> Session info -------------------------------------------------------------
#>  setting  value                       
#>  version  R version 3.4.3 (2017-11-30)
#>  system   x86_64, darwin15.6.0        
#>  ui       X11                         
#>  language (EN)                        
#>  collate  en_US.UTF-8                 
#>  tz       Europe/Madrid               
#>  date     2019-03-15
#> Packages -----------------------------------------------------------------
#>  package   * version    date       source                          
#>  backports   1.1.2      2017-12-13 cran (@1.1.2)                   
#>  base      * 3.4.3      2017-12-07 local                           
#>  compiler    3.4.3      2017-12-07 local                           
#>  crayon      1.3.4      2017-09-16 cran (@1.3.4)                   
#>  datasets  * 3.4.3      2017-12-07 local                           
#>  devtools    1.13.4     2017-11-09 CRAN (R 3.4.2)                  
#>  digest      0.6.18     2018-10-10 cran (@0.6.18)                  
#>  evaluate    0.10.1     2017-06-24 CRAN (R 3.4.0)                  
#>  forcats     0.4.0      2019-02-17 cran (@0.4.0)                   
#>  graphics  * 3.4.3      2017-12-07 local                           
#>  grDevices * 3.4.3      2017-12-07 local                           
#>  haven     * 2.1.0.9000 2019-03-15 Github (tidyverse/haven@a4f3f86)
#>  hms         0.4.2      2018-03-10 cran (@0.4.2)                   
#>  htmltools   0.3.6      2017-04-28 CRAN (R 3.4.0)                  
#>  knitr       1.19       2018-01-29 CRAN (R 3.4.3)                  
#>  magrittr    1.5        2014-11-22 CRAN (R 3.4.0)                  
#>  memoise     1.1.0      2018-09-09 Github (hadley/memoise@06d16ec) 
#>  methods   * 3.4.3      2017-12-07 local                           
#>  pillar      1.3.1      2018-12-15 cran (@1.3.1)                   
#>  pkgconfig   2.0.2      2018-08-16 cran (@2.0.2)                   
#>  Rcpp        1.0.0      2018-11-07 cran (@1.0.0)                   
#>  rlang       0.3.1      2019-01-08 cran (@0.3.1)                   
#>  rmarkdown   1.8        2017-11-17 CRAN (R 3.4.2)                  
#>  rprojroot   1.3-2      2018-01-03 cran (@1.3-2)                   
#>  stats     * 3.4.3      2017-12-07 local                           
#>  stringi     1.2.3      2018-06-12 cran (@1.2.3)                   
#>  stringr     1.3.1      2018-05-10 cran (@1.3.1)                   
#>  tibble      2.0.1      2019-01-12 cran (@2.0.1)                   
#>  tools       3.4.3      2017-12-07 local                           
#>  utils     * 3.4.3      2017-12-07 local                           
#>  withr       2.1.2.9000 2018-12-05 Github (r-lib/withr@be57595)    
#>  yaml        2.1.17     2018-02-27 cran (@2.1.17)

[Gootjes]

I think the issue arises in this defineVariable function: https://github.com/tidyverse/haven/blob/7c24314b7a39f045d95719b21de553bbb2b7c865/src/DfWriter.cpp#L223 The function assumes the existence of a labelset if a variable is of class haven_labelled. Later on in the function, the function is perfectly happy to set labelset to a NULL value. Should be an easy fix: check and allow for a NULL labelset.

…

On Mon, Aug 26, 2019, 05:23 nbchan ***@***.***> wrote: Wrote this function thanks to the workaround suggested by @Gootjes <https://github.com/Gootjes>. Hope someone would find this useful. fix_haven_labelled_write_problem <- function(df){ # Provides a workaround to the problem of writing haven-labelled variables (with null `labels` # and non-null `label`) to Stata / SPSS files. Returns a dataframe compatible with functions # such as `write_sav`. # Reference: #442 # df: dataframe # vector of problematic column names cnames <- df %>% select_if(~ is.labelled(.) & is.null(attributes(.)$labels)) %>% colnames # convert the problematic columns back to their generic data types df_fixed <- df %>% mutate_if(~ is.labelled(.) & is.null(attributes(.)$labels) & typeof(.) == 'character', as.character) %>% mutate_if(~ is.labelled(.) & is.null(attributes(.)$labels) & typeof(.) == 'integer', as.integer) %>% mutate_if(~ is.labelled(.) & is.null(attributes(.)$labels) & typeof(.) == 'double', as.double) # add an attribute called `label` to the respective columns for(cname in cnames){ attributes(df_fixed[[cname]])$label <- df[[cname]] %>% attributes %$% label } return(df_fixed) } Created on 2019-08-26 by the reprex package <https://reprex.tidyverse.org> (v0.3.0) — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#442?email_source=notifications&email_token=AJAX3UZVDDI54FMMHPGXTMDQGNEEVA5CNFSM4G65GXU2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOD5DE2XI#issuecomment-524701021>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AJAX3U22F7X4LASKGKY2ZEDQGNEEVANCNFSM4G65GXUQ> .

[hadley]

Minimal reprex:

library(haven)

x <- labelled(1:3, labels = NULL, label = 'x')
df <- data.frame(x = x)

write_dta(df, tempfile(fileext = '.dta'), version = 14)
#> Error in get_result(output = out, options) : 
#>  callr subprocess failed: could not start R, exited with non-zero status, has crashed or was killed

[lock]

This old issue has been automatically locked. If you believe you have found a related problem, please file a new issue (with reprex) and link to this issue. https://reprex.tidyverse.org/