Error: Column name tidyr::nesting(...) must not be duplicated.... when using complete

[wolski]

After lately updating several packages including tidyr this code stopped working

library(tidyverse)
exdata <- readRDS(url("https://github.com/tidyverse/tidyr/files/3918551/exdata.zip"))
res <- tidyr::complete(
  exdata$data,
  tidyr::nesting(!!!syms(exdata$set1)),
  tidyr::nesting(!!!syms(exdata$set2))
)

and throws the follwing error:
Error: Column name tidyr::nesting(...) must not be duplicated. Use .name_repair to specify repair.

Just changing the order of the columns [c(5,1:4)] in second nesting directive makes it work again (why? - seems to me a spurious change which should not change the behaviour of complete).

res <- tidyr::complete(
  exdata$data,
  tidyr::nesting(!!!syms(exdata$set1)),
  tidyr::nesting(!!!syms(exdata$set2[c(5,1:4)]))
)

> sessionInfo()
R version 3.6.1 (2019-07-05)
Platform: x86_64-w64-mingw32/x64 (64-bit)
Running under: Windows 10 x64 (build 18362)

Matrix products: default

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

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

other attached packages:
[1] forcats_0.4.0   stringr_1.4.0   dplyr_0.8.3     purrr_0.3.3     readr_1.3.1    
[6] tidyr_1.0.0     tibble_2.1.3    ggplot2_3.2.1   tidyverse_1.3.0

loaded via a namespace (and not attached):
 [1] tidyselect_0.2.5 xfun_0.11        haven_2.2.0      lattice_0.20-38 
 [5] colorspace_1.4-1 vctrs_0.2.0      generics_0.0.2   htmltools_0.4.0 
 [9] rlang_0.4.2      pillar_1.4.2     withr_2.1.2      glue_1.3.1      
[13] DBI_1.0.0        dbplyr_1.4.2     readxl_1.3.1     modelr_0.1.5    
[17] lifecycle_0.1.0  cellranger_1.1.0 munsell_0.5.0    gtable_0.3.0    
[21] rvest_0.3.5      evaluate_0.14    knitr_1.26       callr_3.3.2     
[25] ps_1.3.0         broom_0.5.2      Rcpp_1.0.3       clipr_0.7.0     
[29] backports_1.1.5  scales_1.1.0     jsonlite_1.6     fs_1.3.1        
[33] hms_0.5.2        packrat_0.5.0    digest_0.6.23    stringi_1.4.3   
[37] processx_3.4.1   grid_3.6.1       cli_1.1.0        tools_3.6.1     
[41] magrittr_1.5     lazyeval_0.2.2   crayon_1.3.4     whisker_0.4     
[45] pkgconfig_2.0.3  zeallot_0.1.0    ellipsis_0.3.0   xml2_1.2.2      
[49] reprex_0.3.0     lubridate_1.7.4  assertthat_0.2.1 rmarkdown_1.18  
[53] httr_1.4.1       rstudioapi_0.10  R6_2.4.1         nlme_3.1-142    
[57] compiler_3.6.1  

[batpigandme]

Do you mean for there to be two separate tidyr::nesting()s in there? If you use set1 and set2 in a single nesting() you get:

library(tidyverse)
exdata <- readRDS(url("https://github.com/tidyverse/tidyr/files/3918551/exdata.zip"))
res <- tidyr::complete(
  exdata$data,
  tidyr::nesting(!!!syms(exdata$set1), !!!syms(exdata$set2))
)

glimpse(res)
#> Rows: 100
#> Columns: 12
#> $ protein_Id                <chr> "A0AVT1", "A0AVT1", "A0AVT1", "A0AVT1", "A0…
#> $ peptide_Id                <chr> "ACIGDTLCQK", "FISADVHGIWSR", "GMITVTDPDLIE…
#> $ modPeptide_Id             <chr> "_AC[+57]IGDTLC[+57]QK_", "_FISADVHGIWSR_",…
#> $ precursor_Id              <chr> "_AC[+57]IGDTLC[+57]QK_~2", "_FISADVHGIWSR_…
#> $ isotope                   <chr> "light", "light", "light", "light", "light"…
#> $ filename                  <chr> "samon_D1807_098", "samon_D1807_098", "samo…
#> $ sampleName                <chr> "CMP/MEP~patient.HU.yes~PID282", "CMP/MEP~p…
#> $ celltype_                 <chr> "CMP/MEP", "CMP/MEP", "CMP/MEP", "CMP/MEP",…
#> $ class_therapy_progressed_ <fct> patient.HU.yes, patient.HU.yes, patient.HU.…
#> $ Patient_ID                <chr> "PID282", "PID282", "PID282", "PID282", "PI…
#> $ FG.Quantity               <dbl> 104890.961, 63959.547, 56562.879, 7808.923,…
#> $ EG.Qvalue                 <dbl> 9.568321e-04, 1.262353e-04, 5.092613e-04, 8…

^{Created on 2019-12-04 by the reprex package (v0.3.0.9001)}

Adding backtrace for the code as you have it, above.

<error/tibble_error_column_names_must_be_unique>
Column name `tidyr::nesting(...)` must not be duplicated.
Backtrace:
  1. tidyr::complete(...)
  2. tidyr:::complete.data.frame(...)
  4. tidyr:::expand.data.frame(data, ...)
  5. tidyr::expand_grid(!!!cols, .name_repair = .name_repair)
  7. tibble:::as_tibble.list(out, .name_repair = .name_repair)
  8. tibble:::lst_to_tibble(x, .rows, .name_repair, col_lengths(x))
  9. tibble:::set_repaired_names(x, .name_repair)
 14. tibble:::repaired_names(...)
 15. tibble:::subclass_name_repair_errors(...)
 16. base::tryCatch(...)
 17. base:::tryCatchList(expr, classes, parentenv, handlers)
 18. base:::tryCatchOne(...)
 19. value[[3L]](cond)

[wolski]

@batpigandme NO

[batpigandme]

Ok, I'm gonna go ahead and close this issue, then. Let me know if I've misunderstood.

[wolski]

Yes you missunderstood.

[batpigandme]

Ok. Can you please expand on what you mean, then? What's your expected output?

Do you mean for there to be two separate tidyr::nesting()s in there?

@batpigandme No

[wolski]

NO - the issue is not me using two nesting statements! The issue is that the code as posted in the first snipped does not work - and I can't see a reason why it should not work as it is (but this is clearly stated in my first post I believed - and did't I state that it was working with a pre 1.0 version of tidyr?). The expected output is that the first code snipped works, that it does not throw an error (obviously), and that the output is equal (what might be different is the column order) to the result of code snipped 2, which does still work.

[hadley]

@wolski I can tell that you are frustrated but that does not make it ok to be rude. It is hard to tell what you are trying to achieve, and @batpigandme has been trying to help figure out exactly what the problem is.

[wolski]

@hadley This is unfair! I am also trying to help. All I was asking is why code, which I think should work does not. I invested quite a bit of time creating the issue, testing the code, included a reprex and even an example where it works.

[hadley]

@wolski responding "NO" to an attempt to help comes across as very rude. I can see that you have put some time in to this issue, but you have failed to explain it to us in a way we understand. I am going to close this issue to give every one time to cool off; please feel free to open a new one next week.

[wolski]

@hadley The "NO" is rude? "The pot calling the kettle black". Here it was afternoon or evening when I started to reply to @batpigandme question. I was alone at home with two kids 2 and 7 years old. They needed something, so I saved my reply. Ten minutes later when I get back to the issue I see that the issue which I spend several hours preparing is closed. I find the time to say that it is misunderstanding and later a bit more time to answer. And than I am being called rude twice. I was not but now you got me frustrated. For sure I am not going to reopen the issue.