col_types is processed differently by read_*() and type_convert()

[chainsawriot]

In the documentation, col_types appears to be processed the same way by read_*() and type_convert().

type_convert() cannot process "-" (skip).

readr::read_csv(I("a,1,1\nb,1,2\nc,1,2"), col_names = c("a", "b", "c"), col_types = "f-i")
#> # A tibble: 3 × 2
#>   a         c
#>   <fct> <int>
#> 1 a         1
#> 2 b         2
#> 3 c         2

df <- tibble::tibble(a = c("a", "b", "c"),
                     b = c("1", "1", "1"),
                     c = c("1", "2", "2"))
readr::type_convert(df, col_types = "f-i")
#> Warning: Insufficient `col_types`. Guessing 1 columns.
#> Error in if (is.na(name)) {: argument is of length zero
sessionInfo()
#> R version 4.3.1 (2023-06-16)
#> Platform: x86_64-pc-linux-gnu (64-bit)
#> Running under: Ubuntu 22.04.3 LTS
#> 
#> Matrix products: default
#> BLAS:   /usr/lib/x86_64-linux-gnu/blas/libblas.so.3.10.0 
#> LAPACK: /usr/lib/x86_64-linux-gnu/lapack/liblapack.so.3.10.0
#> 
#> locale:
#>  [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C              
#>  [3] LC_TIME=de_DE.UTF-8        LC_COLLATE=en_US.UTF-8    
#>  [5] LC_MONETARY=de_DE.UTF-8    LC_MESSAGES=en_US.UTF-8   
#>  [7] LC_PAPER=de_DE.UTF-8       LC_NAME=C                 
#>  [9] LC_ADDRESS=C               LC_TELEPHONE=C            
#> [11] LC_MEASUREMENT=de_DE.UTF-8 LC_IDENTIFICATION=C       
#> 
#> time zone: Europe/Berlin
#> tzcode source: system (glibc)
#> 
#> attached base packages:
#> [1] stats     graphics  grDevices utils     datasets  methods   base     
#> 
#> loaded via a namespace (and not attached):
#>  [1] crayon_1.5.2      vctrs_0.6.3       cli_3.6.1         knitr_1.43       
#>  [5] rlang_1.1.1       xfun_0.40         purrr_1.0.2       styler_1.10.1    
#>  [9] bit_4.0.5         glue_1.6.2        htmltools_0.5.6   hms_1.1.3        
#> [13] fansi_1.0.4       rmarkdown_2.24    R.cache_0.16.0    evaluate_0.21    
#> [17] tibble_3.2.1      tzdb_0.4.0        fastmap_1.1.1     yaml_2.3.7       
#> [21] lifecycle_1.0.3   compiler_4.3.1    fs_1.6.3          pkgconfig_2.0.3  
#> [25] R.oo_1.25.0       R.utils_2.12.2    digest_0.6.33     R6_2.5.1         
#> [29] tidyselect_1.2.0  utf8_1.2.3        reprex_2.0.2      readr_2.1.4      
#> [33] parallel_4.3.1    vroom_1.6.3       pillar_1.9.0      magrittr_2.0.3   
#> [37] R.methodsS3_1.8.2 bit64_4.0.5       tools_4.3.1       withr_2.5.0

^{Created on 2023-08-22 with reprex v2.0.2}

[jennybc]

Once readr 2e was introduced, which basically means that by default readr is a front end for vroom, there is no longer any real connection between type_convert() (readr 1e code) and the code hit via read_*() (readr 2e = vroom). And thus small discrepancies start to creep in like this.