`tibble 3.0.0` (on CRAN) breaks a number of glance methods

[IndrajeetPatil]

I think this has to do with tibble 3.0.0:

# setup
set.seed(123)
library(robustbase)
library(broom)

# dataframe
data(coleman)
clotting <- data.frame(
  u = c(5, 10, 15, 20, 30, 40, 60, 80, 100),
  lot1 = c(118, 58, 42, 35, 27, 25, 21, 19, 18),
  lot2 = c(69, 35, 26, 21, 18, 16, 13, 12, 12)
)


mod1 <- robustbase::glmrob(
      formula = lot1 ~ log(u),
      data = clotting,
      family = Gamma
    )

mod2 <- robustbase::lmrob(formula = Y ~ ., data = coleman)

glance(mod1)
#> Error: Input must be a vector, not NULL.
glance(mod2)
#> # A tibble: 1 x 3
#>   r.squared sigma df.residual
#>       <dbl> <dbl>       <int>
#> 1     0.981  1.13          14

[IndrajeetPatil]

Trace for this error:

> rlang::last_trace()
<error/vctrs_error_scalar_type>
Input must be a vector, not NULL.
Backtrace:
     x
  1. +-(function (x, ...) ...
  2. +-tibble:::print.tbl(x)
  3. | +-tibble:::cat_line(format(x, ..., n = n, width = width, n_extra = n_extra))
  4. | | +-base::cat(paste0(..., "\n"), sep = "")
  5. | | \-base::paste0(..., "\n")
  6. | +-base::format(x, ..., n = n, width = width, n_extra = n_extra)
  7. | \-tibble:::format.tbl(x, ..., n = n, width = width, n_extra = n_extra)
  8. |   \-tibble::trunc_mat(x, n = n, width = width, n_extra = n_extra)
  9. |     +-base::as.data.frame(head(x, n))
 10. |     +-utils::head(x, n)
 11. |     \-utils:::head.data.frame(x, n)
 12. |       +-base::do.call("[", args)
 13. |       +-x[integer(0), , drop = FALSE]
 14. |       \-tibble:::`[.tbl_df`(x, integer(0), , drop = FALSE)
 15. |         \-tibble:::tbl_subset_row(xo, i = i)
 16. |           \-base::lapply(unclass(x), vec_slice, i = i)
 17. |             \-vctrs:::FUN(X[[i]], ...)
 18. \-vctrs:::stop_scalar_type(.Primitive("quote")(NULL), "")
 19.   \-vctrs:::stop_vctrs(msg, "vctrs_error_scalar_type", actual = x)

[IndrajeetPatil]

Another example with bracl

# setup
set.seed(123)
library(brglm2)
data("stemcell")

# bias reduction for adjacent category logit models
# for ordinal responses using the Poisson trick
fit_bracl <-
  brglm2::bracl(
    formula = research ~ as.numeric(religion) + gender,
    weights = frequency,
    data = stemcell,
    type = "ML"
  )

# glance
broom::glance(fit_bracl)
#> Error: Input must be a vector, not NULL.
#> Backtrace:
#>      x
#>   1. +-base::tryCatch(...)
#>   2. | \-base:::tryCatchList(expr, classes, parentenv, handlers)
#>   3. |   +-base:::tryCatchOne(...)
#>   4. |   | \-base:::doTryCatch(return(expr), name, parentenv, handler)
#>   5. |   \-base:::tryCatchList(expr, names[-nh], parentenv, handlers[-nh])
#>   6. |     \-base:::tryCatchOne(expr, names, parentenv, handlers[[1L]])
#>   7. |       \-base:::doTryCatch(return(expr), name, parentenv, handler)
#>   8. +-base::withCallingHandlers(...)
#>   9. +-base::saveRDS(...)
#>  10. +-base::do.call(...)
#>  11. +-(function (what, args, quote = FALSE, envir = parent.frame()) ...
#>  12. +-(function (input, opts) ...
#>  13. | \-rmarkdown::render(input, quiet = TRUE, envir = globalenv(), encoding = "UTF-8")
#>  14. |   \-knitr::knit(knit_input, knit_output, envir = envir, quiet = quiet)
#>  15. |     \-knitr:::process_file(text, output)
#>  16. |       +-base::withCallingHandlers(...)
#>  17. |       +-knitr:::process_group(group)
#>  18. |       \-knitr:::process_group.block(group)
#>  19. |         \-knitr:::call_block(x)
#>  20. |           \-knitr:::block_exec(params)
#>  21. |             +-knitr:::in_dir(...)
#>  22. |             \-knitr:::evaluate(...)
#>  23. |               \-evaluate::evaluate(...)
#>  24. |                 \-evaluate:::evaluate_call(...)
#>  25. |                   +-evaluate:::handle(...)
#>  26. |                   | \-base::try(f, silent = TRUE)
#>  27. |                   |   \-base::tryCatch(...)
#>  28. |                   |     \-base:::tryCatchList(expr, classes, parentenv, handlers)
#>  29. |                   |       \-base:::tryCatchOne(expr, names, parentenv, handlers[[1L]])
#>  30. |                   |         \-base:::doTryCatch(return(expr), name, parentenv, handler)
#>  31. |                   +-base::withCallingHandlers(...)
#>  32. |                   +-base::withVisible(value_fun(ev$value, ev$visible))
#>  33. |                   \-knitr:::value_fun(ev$value, ev$visible)
#>  34. |                     \-knitr:::fun(x, options = options)
#>  35. |                       +-base::withVisible(knit_print(x, ...))
#>  36. |                       +-knitr::knit_print(x, ...)
#>  37. |                       \-rmarkdown:::knit_print.data.frame(x, ...)
#>  38. |                         +-context$df_print(x)
#>  39. |                         \-tibble:::print.tbl(x)
#>  40. |                           +-cli::cat_line(format(x, ..., n = n, width = width, n_extra = n_extra))
#>  41. |                           | \-base::paste0(..., collapse = "\n")
#>  42. |                           +-base::format(x, ..., n = n, width = width, n_extra = n_extra)
#>  43. |                           \-tibble:::format.tbl(x, ..., n = n, width = width, n_extra = n_extra)
#>  44. |                             \-tibble::trunc_mat(x, n = n, width = width, n_extra = n_extra)
#>  45. |                               +-base::as.data.frame(head(x, n))
#>  46. |                               +-utils::head(x, n)
#>  47. |                               \-utils:::head.data.frame(x, n)
#>  48. |                                 +-base::do.call("[", args)
#>  49. |                                 +-x[1L, , drop = FALSE]
#>  50. |                                 \-tibble:::`[.tbl_df`(x, 1L, , drop = FALSE)
#>  51. |                                   \-tibble:::tbl_subset_row(xo, i = i, i_arg)
#>  52. |                                     \-base::lapply(unclass(x), vec_slice, i = i)
#>  53. |                                       \-vctrs:::FUN(X[[i]], ...)
#>  54. \-vctrs:::stop_scalar_type(.Primitive("quote")(NULL), "")
#>  55.   \-vctrs:::stop_vctrs(msg, "vctrs_error_scalar_type", actual = x)

[alexpghayes]

At the moment I'm a bit overwhelmed. I'll try to get to this when I can, but my priority at the moment is to get the CRAN version of broom to be happy with tibble 3.0.0. It may be a moment.

[IndrajeetPatil]

@simonpcouch Can this be also resolved before the 0.7.0 release?

[simonpcouch]

Moving over to this issue in the next couple days! Thanks for checking in!

[simonpcouch]

Seems like the issue here is that tibble will now error, rather than returning a 0-row tibble / tibble with NULLs, when given NULL as a column entry. When a fit is rank-deficient(?), many of the entries dropped into glance.*() tibbles are NULL. Working on a helper right now that, given a named list (supplied in ...), will return a 1-row tibble with appropriate NA types switched out for NULL values.

[github-actions]

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