revdep issues caused when . and J are exported

[tdhock]

Revdeps affected

By .

• animint2 (2024.1.24, 2024-01)
• AFM (2.0, 2020-10)
• Anaconda (0.1.5, 2022-10)
• EurosarcBayes (1.1, 2017-11)
• FAOSTAT (2.4.0, 2024-05)
• FMAT (2024.6, 2024-06)
• Fgmutils (0.9.5, 2018-11)
• PTXQC (1.1.1, 2024-03)
• antaresRead (2.7.0, 2024-05)
• dfmeta (1.0.0, 2018-03)
• distantia (1.0.2, 2019-10)
• eyeTrackR (1.0.1, 2020-03)
• growthcleanr (2.2.0, 2024-02)
• llama (0.10.1, 2021-03)
• lwqs (0.5.0, 2021-03)
• pda (1.2.7, 2024-03)
• rqPen (4.1.1, 2024-06)

By J

• llama (0.10.1, 2021-03)

---

#6125 caused new check issues with 17 revdeps, mostly because the . is also exported by plyr. also rJava::J in one package.
Most fixes should be easy, revdeps should use importFrom instead of import in their NAMESPACE.
Should I go ahead and start contacting revdeps? or should we consider reverting #6125 ?

animint2

* checking Rd files ... WARNING
prepare_Rd: replacing previous import 'data.table::.' by 'plyr::.' when loading 'animint2'

* checking whether package 'animint2' can be installed ... WARNING
Found the following significant warnings:
  Warning: replacing previous import 'data.table::.' by 'plyr::.' when loading 'animint2'

AFM

* checking whether package 'AFM' can be installed ... WARNING
Found the following significant warnings:
  Warning: replacing previous import 'data.table::.' by 'plyr::.' when loading 'AFM'

Anaconda

* checking whether package 'Anaconda' can be installed ... WARNING
Found the following significant warnings:
  Warning: replacing previous import 'data.table::.' by 'plyr::.' when loading 'Anaconda'

EurosarcBayes

* checking whether package 'EurosarcBayes' can be installed ... WARNING
Found the following significant warnings:
  Warning: replacing previous import 'plyr::.' by 'data.table::.' when loading 'EurosarcBayes'

FAOSTAT

* checking whether package 'FAOSTAT' can be installed ... WARNING
Found the following significant warnings:
  Warning: replacing previous import 'data.table::.' by 'plyr::.' when loading 'FAOSTAT'

Fgmutils

* checking whether package 'Fgmutils' can be installed ... WARNING
Found the following significant warnings:
  Warning: replacing previous import 'data.table::.' by 'plyr::.' when loading 'Fgmutils'

PTXQC

* checking whether package 'PTXQC' can be installed ... WARNING
Found the following significant warnings:
  Warning: replacing previous import 'data.table::.' by 'plyr::.' when loading 'PTXQC'

antaresRead

* checking whether package 'antaresRead' can be installed ... WARNING
Found the following significant warnings:
  Warning: replacing previous import 'data.table::.' by 'plyr::.' when loading 'antaresRead'

dfmeta

* checking whether package 'dfmeta' can be installed ... WARNING
Found the following significant warnings:
  Warning: replacing previous import 'data.table::.' by 'plyr::.' when loading 'dfmeta'

distantia

* checking whether package 'distantia' can be installed ... WARNING
Found the following significant warnings:
  Warning: replacing previous import 'data.table::.' by 'plyr::.' when loading 'distantia'

eyeTrackR

* checking whether package 'eyeTrackR' can be installed ... WARNING
Found the following significant warnings:
  Warning: replacing previous import 'data.table::.' by 'plyr::.' when loading 'eyeTrackR'

growthcleanr

* checking whether package 'growthcleanr' can be installed ... WARNING
Found the following significant warnings:
  Warning: replacing previous import 'data.table::.' by 'plyr::.' when loading 'growthcleanr'

llama

* checking whether package 'llama' can be installed ... WARNING
Found the following significant warnings:
  Warning: replacing previous import 'plyr::.' by 'data.table::.' when loading 'llama'
  Warning: replacing previous import 'rJava::J' by 'data.table::J' when loading 'llama'

lwqs

* checking whether package 'lwqs' can be installed ... WARNING
Found the following significant warnings:
  Warning: replacing previous import 'data.table::.' by 'plyr::.' when loading 'lwqs'

pda

* checking whether package 'pda' can be installed ... WARNING
Found the following significant warnings:
  Warning: replacing previous import 'data.table::.' by 'plyr::.' when loading 'pda'

rqPen

* checking whether package 'rqPen' can be installed ... WARNING
Found the following significant warnings:
  Warning: replacing previous import 'data.table::.' by 'plyr::.' when loading 'rqPen'

* checking Rd files ... WARNING
prepare_Rd: replacing previous import 'data.table::.' by 'plyr::.' when loading 'rqPen'

one package has a new error in examples https://github.com/psychbruce/FMAT/issues

* checking examples ... ERROR
Running examples in 'FMAT-Ex.R' failed
The error most likely occurred in:

> base::assign(".ptime", proc.time(), pos = "CheckExEnv")
> ### Name: .
> ### Title: A simple function equivalent to 'list'.
> ### Aliases: .
> 
> ### ** Examples
> 
> .(Male=c("he", "his"), Female=c("she", "her"))
Error in stopf(".() called outside of [.data.table. .() is only intended as an alias for list() inside DT[...].") : 
  .() called outside of [.data.table. .() is only intended as an alias for list() inside DT[...].
Calls: . -> stopf -> raise_condition -> signal
Execution halted

[Nj221102]

Well reverting will retain the issues in #5604 and #5277, wouldn't revdeps using importFrom instead of import in their NAMESPACE will solve both these issues and this new issue as well.

still if its seems like not worth it, then i can make a PR removing masks for . and j as only two of these are causing issues.

[MichaelChirico]

I think there's no rush to export the masks. let's revert . and J, then give these downstreams a very long heads up (>1 year) about our intent to export the masks before proceeding.

[MichaelChirico]

i can make a PR removing masks for . and j as only two of these are causing issues.

Yes, let's do that for 1.16.0. We'll target fixing the other two issues in 1.18.0 (at least one year from 1.16.0).

[MichaelChirico]

@tdhock an aside: it would be nice to link the revdep results at the top of these issues for easier reference

[MichaelChirico]

This is fixed by #6198.

[tdhock]

ok I can link revdep results next time but the links expire after one week so I tend to prefer just copying the relevant info to issues

[MichaelChirico]

Makes sense -- to clarify, I don't need the specific results linked, but just a link to the revdep service itself would be enough. I find myself searching around for the exact URL each time.