Impossible to install in R 4.0 : code error in file cb_summary_stats_numeric.R

[aeshjalmur]

Attempting installation of codebooker in R 4.0 results in the following error which aborts installation:

cb_summary_stats_numeric.R:30:54: unexpected input 29: # Format output 30: dplyr::mutate(dplyr::across(dplyr::everything(), \ ^ ERROR: unable to collate and parse R files for package 'codebookr'

• removing 'D:/Logiciels/R-4.0.0/library/codebookr'

The affected lines of file cb_summary_stats_numeric.R are:

29 # Format output 30 dplyr::mutate(dplyr::across(dplyr::everything(), (x) round(x, digits = digits))) %>% 31 dplyr::mutate(dplyr::across(dplyr::everything(), (x) format(x, nsmall = digits)))

The dash syntax seems to be incompatible with R

Replacement by the following lines solved the installation issue, but I do not know yet if the package will remain functional after this substitution : 29 # Format output 30 dplyr::mutate(dplyr::across(dplyr::everything(), ~ round(.x, digits = digits))) %>% 31 dplyr::mutate(dplyr::across(dplyr::everything(), ~ format(.x, nsmall = digits)))

Best

[mbcann01]

@aeshjalmur Thank you for letting me know. Do you mind telling me what version of R you are using?

[aeshjalmur]

I am using R version 4.0.0 (2020-04-24) -- "Arbor Day" Copyright (C) 2020 The R Foundation for Statistical Computing Platform: x86_64-w64-mingw32/x64 (64-bit).

Additionnally, I tried the codebook function from the package, and it did not work for me as it displayed the following error which I could not solve: Erreur : 'reframe' is not an exported object from 'namespace:dplyr'

[mbcann01]

Hi @aeshjalmur, That makes sense.

• Base R didn't introduce the shorthand syntax for anonymous functions until version 4.1.0. That explains why the code in cb_summary_stats_numeric doesn't work on your computer.
• Similarly, the reframe() function was added to the dplyr package in version 1.1.0. If you are using an older version of dplyr, that would explain the "Erreur : 'reframe' is not an exported object from 'namespace:dplyr'".

You should be able to fix these errors by updating base R and dplyr. However, you may have a good reason for not wanting to do so. On my end, I think I can, and should, update the dependency file to indicate that base R version 4.1 or greater and dplyr version 1.1 or greater are required by this package.