Data.table: testthat does not recognize data.tables ?

[fruce-ki]

I started developing a new package using newer dat.table version and got stopped fairly early on but what seems to be an issue between 'testthat' and 'data.table'. I did not encounter that issue working with both packages for my previous package.

I have a function like this in my package:

#' Title
#' 
#' @import data.table
#' @export
#' 
DT_test <- function()
{
    dt <- data.table(acol= c("aA","aB", "aC"), 
                   bcol= c("bA", "bB", "bC"))
    dt[, ccol := c(1,2,3) ]
    return(dt)
}

And this is the corresponding unittest:

test_that("The data.table assignment works", {
    dt <- DT_test()
       # test stuff
})

I get this error:

Failed -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
1. Error: The data.table assignment works (@test-1-devel.R#17) -----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Check that is.data.table(DT) == TRUE. Otherwise, := and `:=`(...) are defined for use in j, once only and in particular ways. See help(":=").
1: DT_test() at /Users/cymon/workspace/Treat/tests/testthat/test-1-devel.R:17
2: dt[, `:=`(ccol, c(1, 2, 3))] at /Users/cymon/workspace/Treat/R/func.R:53
3: `[.data.table`(dt, , `:=`(ccol, c(1, 2, 3))) at /Users/cymon/workspace/Treat/R/func.R:53
4: `[.data.frame`(x, i, j)
5: `:=`(ccol, c(1, 2, 3))
6: stop("Check that is.data.table(DT) == TRUE. Otherwise, := and `:=`(...) are defined for use in j, once only and in particular ways. See help(\":=\").")

But if I call the function from a prompt or run it line for line it runs fine.

Session info -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
 setting  value                       
 version  R version 3.3.2 (2016-10-31)
 system   x86_64, darwin13.4.0        
 ui       RStudio (1.0.136)           
 language (EN)                        
 collate  en_GB.UTF-8                 
 tz       Europe/London               
 date     2017-03-08                  

Packages -----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
 package    * version date       source        
 commonmark   1.2     2017-03-01 CRAN (R 3.3.2)
 crayon       1.3.2   2016-06-28 CRAN (R 3.3.0)
 data.table * 1.10.4  2017-02-01 CRAN (R 3.3.2)
 devtools   * 1.12.0  2016-06-24 CRAN (R 3.3.0)
 digest       0.6.12  2017-01-27 CRAN (R 3.3.2)
 magrittr     1.5     2014-11-22 CRAN (R 3.3.0)
 memoise      1.0.0   2016-01-29 CRAN (R 3.3.0)
 R6           2.2.0   2016-10-05 CRAN (R 3.3.0)
 Rcpp         0.12.9  2017-01-14 CRAN (R 3.3.2)
 roxygen2     6.0.1   2017-02-06 CRAN (R 3.3.2)
 stringi      1.1.2   2016-10-01 CRAN (R 3.3.0)
 stringr      1.2.0   2017-02-18 CRAN (R 3.3.2)
 testthat   * 1.0.2   2016-04-23 CRAN (R 3.3.0)
 treat      * 0.0.1   <NA>       local         
 withr        1.0.2   2016-06-20 CRAN (R 3.3.0)
 xml2         1.1.1   2017-01-24 CRAN (R 3.3.2)

[fruce-ki]

It seems to be connected to building the package:

• Using devtools::load_all() to load code edits causes the function to throw the same error even outside of testthat.
• Building the package instead, using

devtools::document(roclets=c('rd', 'collate', 'namespace'))
R CMD INSTALL --no-multiarch --with-keep.source Treat

allows the function run correctly, except in unittests (I'm guessing testthat uses load_all())

PS. So far the minimal example is virtually the only thing I have actually implemented in the package. There is no other code that could be interfering or altering behaviour etc.

[jangorecki]

@fruce-ki any idea if there is something we can do about it? Similar issue was discussed very recently, you may want to take a look #2023

[fruce-ki]

@jangorecki It seems to be the same issue, discovered via different route. I don't know if it is because of something data.tables has changed or because of something that devtools has changed. At least it seems it is not something testthat did.

So I guess the next step is to go knock on devtools' door to see if they know anything about what's going on here.

[arunsrinivasan]

Assuming you're importing data.table, do you have import(data.table) in your NAMESPACE.. a.k.a. FAQ 6.9.

I see that you've added it in your roxygen comments.

[chr1swallace]

I had a similar issue. Googling, I found this https://stackoverflow.com/questions/13106018/data-table-error-when-used-through-knitr-gwidgetswww/13131555#13131555
and added

assignInNamespace("cedta.override",
                  c(data.table:::cedta.override,"<my-package-name>"),
                  "data.table")

to the top of the file causing the problem, and it now works. I don't understand why though

[asantucci]

I am having the same problem, but the workaround posted above does not reliably work for me.

This is mildly frustrating. Data.table is in general absolutely fantastic! Any updates on this issue?

[chinsoon12]

would adding library(data.table) in tests/testthat.R help?
couldnt replicate problem on Windows if that helps.

[asantucci]

To no avail, any other suggestions? I really appreciate your time. Please let me know if I can help to provide more information.

[chinsoon12]

throwing random ideas for trying:
try adding requireNamespace("data.table") to the package .onAttach function?
add data.table to both Imports and Suggests in DESCRIPTION?
just build and check in command line directly?

[jangorecki]

@asantucci Be sure to read FAQ about this. You need to have proper import in DESCRIPTION and NAMESPACE files. Do not rely on helpers like roxygen when verifying if your imports are properly defined. Always check if problem still occurs when building and checking from console.

[jangorecki]

If problems cannot be reproduced from R console I think this issue can be closed, unless someone has some action points related to this issue. I found the following issues related: devtools#192 and devtools#1472.

[asantucci]

Thanks, @jangorecki for the related links. For what it's worth, the first, devtools#192, simply dead-ends without a solution (in fact pointing back to this thread). The second, devtools#1472, was almost helpful, in that it encouraged me to re-install/update the r-lib/pkgload such that I could make sure I had the freshest version of devtools::load_all() before trying devtools::test(). Anyway, it's all moot since none of this ends up working. Forgive me if I have missed an obvious solution that's contained within these links...With respect to the FAQ that you mentioned in your previous post, I'd note that it's possible to go "above and beyond" by, e.g., linking to the FAQ and citing which section you're referring to. For what it's worth, this FAQ hasn't popped up in any of my searches along the way.

For using data.table in a package, it's sufficient to declare it as a depends in the DESCRIPTION file.
It turns out that adding
import(data.table)
manually to the NAMESPACE file is also necessary to have it work with testthat. It's a bit lame since Roxygen edits this file each time the package is re-built, meaning we have to manually re-insert this import(data.table) line into the generated file each time we modify our package.

Continually surprised that given the popularity of data.table and testthat, that this is still an ongoing issue. Posting here not because I'm complaining (both packages are individually amazing), but because I'd like for future programmers to find this more easily.

[chinsoon12]

Hi @asantucci, would adding #’ @import data.table in at lease one of the functions that depend on data.table help?

[HughParsonage]

For using data.table in a package, it's sufficient to declare it as a depends in the DESCRIPTION file.

Sufficient, yes. But not necessary.

It turns out that adding
import(data.table)
manually to the NAMESPACE file is also necessary to have it work with testthat. It's a bit lame since Roxygen edits this file each time the package is re-built, meaning we have to manually re-insert this import(data.table) line into the generated file each time we modify our package.

Not my experience. It's possible you edited the NAMESPACE file once, after which roxygen2 can stop modifying it. As chinsoon12 said, adding #' @import data.table is sufficient for data.table to work in a package without requiring data.table be attached (which is what Depends: does).

I suspect that there's something askew in your package: the roxygen roclets, the NAMESPACE file, or the build process itself. My first test would be to remove data.table from Depends: and place it in Imports:, then add #' @import data.table in a file in your R directory, remove your NAMESPACE file, re-document your package with roxygen, then test.

[jangorecki]

@chinsoon12 please read new doc on that: https://rdatatable.gitlab.io/data.table/library/data.table/doc/datatable-importing.html

[asantucci]

I went ahead and did the following

• Removed data.table from Depends
• Removed NAMESPACE file by hand
• Rebuilt documentation
  Still no luck.
  Went ahead and added #' @import data.table to a single function, and suddenly things appeared to "work for a while". Several package rebuilds later, however, I started getting more error messages indicating that data.table was no longer properly being imported by the package. Adding in data.table back into Depends remedies the problem...

@jangorecki Perhaps consider mentioning that this #' @imports data.table may be helpful to some users in the article you linked to on datatable-importing?

Still to be improved (call me greedy); a future developer may stumble upon the lone #' @imports data.table and erroneously remove it. If they don't have the best testing habits, they may remove it, rebuild the package successfully, forget to re-test, then only realize sometime later that their tests are now broken and they can't figure out why. We could add a comment to the #' @imports data.table, but this feels like adding a bunch of band-aids rather than fixing the root of the problem.

Thanks everyone, for your time.

[jangorecki]

@asantucci IMO proper way is to write NAMESPACE by hand :) I am pretty sure roxygen documentation mention what you are asking about. I used roxygen in many packages but only for Rd files, not NAMESPACE file which I like to format well myself. Troubleshooting section in our vignette is quite clear:

Some of the most common issues developers are facing are usually related to helpers tools that meant to automate some package development tasks. For example using roxygen to generate NAMESPACE file from metadata in R code files. Others are related to helpers that build and check package. Thus be sure to double check using R console, also ensure proper import is defined in DESCRIPTION and NAMESPACE files.

[asantucci]

@jangorecki That's great you like to spend time formatting the NAMESPACE by hand. I'd venture a guess that you're a small minority in that regard. I respectfully disagree that I'm "asking about roxygen documentation"; I'm using both Matrix and data.table in the same package and in the same set of tests and while both behave similarly for #' @import or Imports in DESCRIPTION, only one is causing grief in testing :-) Lastly, consider re-reading the "quite clear" vignette; you may realize that it's mantra is of the form "your problems are caused by a general class of widely used tools. Here's mention of one such tool that is popularly used. Please fix the problem by, well, properly using DESCRIPTION and NAMESPACE". Here, I laugh, since this last sentence begs the question.
Again, really appreciative of your time. Simply trying to help others along the way (package maintainers included :-)).
Feel free to view me like this: https://imgur.com/gallery/X17puIB

[mattdowle]

@asantucci Everyone is frustrated.

Not sure if you found this one or not (I can't see it linked above) :
https://stackoverflow.com/a/10529888/403310

That contains the link to the FAQ, too, where the same info is repeated. Searching the FAQ for "import" would have found it, at the point Jan mentioned there was a FAQ about it. (Maybe he was writing from a phone which is why he didn't give you a link directly to it.)

The fact is that Roxygen/Rstudio attempts to write and overwrite your NAMESPACE file for you, to make your life easier. But sometimes fails, as in this case it seems, and makes your life harder. Please talk to them about that and let us know how you get on.

If an imported package does not actually appear in NAMESPACE, then it isn't going to work. So while this apparent bug in Roxygen/Rstudio persists, you need to carefully check your NAMESPACE file to see if that line is there or not.

Another annoyance is that R needs the imported package to be named in both places: DESCRIPTION and NAMESPACE. And if it is in only one of those, there is no helpful warning or note as far as I remember. It could be that recent versions of R now warn/note and it could be that RStudio does not show that to you. It all takes time to investigate, search r-devel history and liaise with people. Since it effects you most, could you do that please? An update to R, or an update to roxygen/RStudio in this regard would be appreciated by a great many folk. As far as I know it affects a lot of packages and is nothing to do with data.table per se.

[ChandlerLutz]

@mattdowle and @jangorecki Thanks for this. Just to summarize:

See for reference:

https://stackoverflow.com/a/10529888/403310

https://stackoverflow.com/a/36930626/1317443

https://rdatatable.gitlab.io/data.table/library/data.table/doc/datatable-importing.html

I was able to solve these issues with roxygen2 v6.1 data.table 1.11.4 and R 3.5.1 by

1. using devtools::use_package(data.table, type = "depends")
2. making sure that data.table is imported in NAMESPACE by putting @import data.table in my main package documentation

[douglas-hoen]

Thank you all for this. I had the same issue. After much frustration, as @ChandlerLutz and others stated above, simply adding #' @import data.table above a function in my main package (/R folder not /tests) solved it for me.

(I made an empty function just for this purpose and explicitly documented it, so that it will not likely be later removed by accident. NB: I also have data.table in the Imports section of my DESCRIPTION file.)

R 3.5.2, data.table 1.12.0, roxygen2 6.1.0.9000

[tdhock]

hi I am having this issue as well, but my package suggests data.table (not imports). I reported the issue to testthat https://github.com/r-lib/testthat/issues/933 but @hadley wrote "This is almost certainly because of the way that data.table checks for whether or not it's in a data.table aware environment, and is unlikely to be directly related to testthat"

to reproduce the issue I created a MRE package here, https://github.com/tdhock/testdtbug

In particular the [.data.frame method seems to be incorrectly called when using [ on a data.table during test_package (but not test_file), which results in an "object not found" error.

I expected that tests should pass using both test_package and test_file, but tests only pass using test_file.

the code to reproduce the error is:

remotes::install_github("tdhock/testdtbug", force=TRUE, INSTALL_opts="--install-tests")
testthat::test_package("testdtbug")
testthat::test_file(system.file("tests", "testthat", "test-dt.R", package="testdtbug"))
devtools::session_info()

The output on my system was:

> remotes::install_github("tdhock/testdtbug", force=TRUE, INSTALL_opts="--install-tests")
Downloading GitHub repo tdhock/testdtbug@master
v  checking for file 'C:\Users\th798\AppData\Local\Temp\RtmpK2I3xH\remotes33186a156572\tdhock-testdtbug-e7a14c5/DESCRIPTION'
-  preparing 'testdtbug':
v  checking DESCRIPTION meta-information
-  checking for LF line-endings in source and make files and shell scripts
-  checking for empty or unneeded directories
-  creating default NAMESPACE file
-  building 'testdtbug_2019.9.3.tar.gz'

Installing package into 'C:/Users/th798/R/win-library/3.6'
(as 'lib' is unspecified)
* installing *source* package 'testdtbug' ...
** using staged installation
** tests
** help
No man pages found in package  'testdtbug' 
*** installing help indices
** building package indices
** testing if installed package can be loaded from temporary location
** testing if installed package can be loaded from final location
** testing if installed package keeps a record of temporary installation path
* DONE (testdtbug)
> testthat::test_package("testdtbug")
-- 1. Error: dt [ method ok (@test-dt.R#5)  ------------------------------------
object 'foo' not found
1: data.table(group = 1:2, foo = "bar")[, {
       expect_identical(foo, "bar")
   }, by = group] at C:/Users/th798/R/win-library/3.6/testdtbug/tests/testthat/test-dt.R:5
2: `[.data.table`(data.table(group = 1:2, foo = "bar"), , {
       expect_identical(foo, "bar")
   }, by = group) at C:/Users/th798/R/win-library/3.6/testdtbug/tests/testthat/test-dt.R:5
3: `[.data.frame`(x, i, j)
4: expect_identical(foo, "bar") at C:/Users/th798/R/win-library/3.6/testdtbug/tests/testthat/test-dt.R:6
5: quasi_label(enquo(object), label, arg = "object")
6: eval_bare(get_expr(quo), get_env(quo))

== testthat results  ===========================================================
[ OK: 0 | SKIPPED: 0 | WARNINGS: 0 | FAILED: 1 ]
1. Error: dt [ method ok (@test-dt.R#5) 

Error: testthat unit tests failed
> testthat::test_file(system.file("tests", "testthat", "test-dt.R", package="testdtbug"))
v |  OK F W S | Context
v |   2       | dt

== Results =====================================================================
OK:       2
Failed:   0
Warnings: 0
Skipped:  0
> devtools::session_info()
- Session info ---------------------------------------------------------------
 setting  value                       
 version  R version 3.6.1 (2019-07-05)
 os       Windows 10 x64              
 system   x86_64, mingw32             
 ui       RTerm                       
 language (EN)                        
 collate  English_United States.1252  
 ctype    English_United States.1252  
 tz       America/Phoenix             
 date     2019-09-04                  

- Packages -------------------------------------------------------------------
 ! package      * version   date       lib source                           
   assertthat     0.2.1     2019-03-21 [1] CRAN (R 3.6.1)                   
   backports      1.1.4     2019-04-10 [1] CRAN (R 3.6.1)                   
   callr          3.3.1     2019-07-18 [1] CRAN (R 3.6.1)                   
   cli            1.1.0     2019-03-19 [1] CRAN (R 3.6.1)                   
   crayon         1.3.4     2017-09-16 [1] CRAN (R 3.6.1)                   
   curl           3.3       2019-01-10 [1] CRAN (R 3.6.1)                   
   data.table   * 1.12.2    2019-04-07 [1] standard (@1.12.2)               
   desc           1.2.0     2018-05-01 [1] CRAN (R 3.6.1)                   
   devtools       2.1.0     2019-07-06 [1] CRAN (R 3.6.1)                   
   digest         0.6.20    2019-07-04 [1] CRAN (R 3.6.1)                   
   fs             1.3.1     2019-05-06 [1] CRAN (R 3.6.1)                   
   glue           1.3.1     2019-03-12 [1] CRAN (R 3.6.1)                   
   htmltools      0.3.6     2017-04-28 [1] CRAN (R 3.6.1)                   
   htmlwidgets    1.3       2018-09-30 [1] CRAN (R 3.6.1)                   
   magrittr       1.5       2014-11-22 [1] CRAN (R 3.6.1)                   
   memoise        1.1.0     2017-04-21 [1] CRAN (R 3.6.1)                   
   namedCapture * 2019.8.13 2019-08-14 [1] local                            
   pkgbuild       1.0.3     2019-03-20 [1] CRAN (R 3.6.1)                   
   pkgload        1.0.2     2018-10-29 [1] CRAN (R 3.6.1)                   
   prettyunits    1.0.2     2015-07-13 [1] CRAN (R 3.6.1)                   
   processx       3.4.1     2019-07-18 [1] CRAN (R 3.6.1)                   
   ps             1.3.0     2018-12-21 [1] CRAN (R 3.6.1)                   
   R6             2.4.0     2019-02-14 [1] CRAN (R 3.6.1)                   
   Rcpp           1.0.2     2019-07-25 [1] CRAN (R 3.6.1)                   
 D RcppParallel   4.4.3     2019-05-22 [1] CRAN (R 3.6.1)                   
   re2r           0.2.0     2017-09-04 [1] CRAN (R 3.6.1)                   
   remotes        2.1.0     2019-06-24 [1] standard (@2.1.0)                
   rlang          0.4.0     2019-06-25 [1] CRAN (R 3.6.1)                   
   rprojroot      1.3-2     2018-01-03 [1] CRAN (R 3.6.1)                   
   sessioninfo    1.1.1     2018-11-05 [1] CRAN (R 3.6.1)                   
   stringi        1.4.3     2019-03-12 [1] CRAN (R 3.6.0)                   
   testdtbug      2019.9.3  2019-09-04 [1] Github (tdhock/testdtbug@e7a14c5)
   testthat     * 2.2.0     2019-07-22 [1] CRAN (R 3.6.1)                   
   usethis        1.5.1     2019-07-04 [1] CRAN (R 3.6.1)                   
   withr          2.1.2     2018-03-15 [1] CRAN (R 3.6.1)                   

[1] C:/Users/th798/R/win-library/3.6
[2] c:/Program Files/R/R-3.6.1/library

 D -- DLL MD5 mismatch, broken installation.
> 

This also happens when running R --vanilla:

> remotes::install_github('tdhock/testdtbug', force=TRUE, INSTALL_opts='--install-tests')
Downloading GitHub repo tdhock/testdtbug@master
v  checking for file 'C:\Users\th798\AppData\Local\Temp\RtmpCCwTzT\remotes2804fa96872\tdhock-testdtbug-e7a14c5/DESCRIPTION'
-  preparing 'testdtbug':
v  checking DESCRIPTION meta-information
-  checking for LF line-endings in source and make files and shell scripts
-  checking for empty or unneeded directories
-  creating default NAMESPACE file
-  building 'testdtbug_2019.9.3.tar.gz'

Installing package into 'C:/Users/th798/R/win-library/3.6'
(as 'lib' is unspecified)
* installing *source* package 'testdtbug' ...
** using staged installation
** tests
** help
No man pages found in package  'testdtbug' 
*** installing help indices
** building package indices
** testing if installed package can be loaded from temporary location
** testing if installed package can be loaded from final location
** testing if installed package keeps a record of temporary installation path
* DONE (testdtbug)
> 
> 
]0;MINGW64:/c/Users/th798/R
th798@cmp2986 MINGW64 ~/R
$ R --vanilla -e "testthat::test_package('testdtbug')"

R version 3.6.1 (2019-07-05) -- "Action of the Toes"
Copyright (C) 2019 The R Foundation for Statistical Computing
Platform: x86_64-w64-mingw32/x64 (64-bit)

R is free software and comes with ABSOLUTELY NO WARRANTY.
You are welcome to redistribute it under certain conditions.
Type 'license()' or 'licence()' for distribution details.

  Natural language support but running in an English locale

R is a collaborative project with many contributors.
Type 'contributors()' for more information and
'citation()' on how to cite R or R packages in publications.

Type 'demo()' for some demos, 'help()' for on-line help, or
'help.start()' for an HTML browser interface to help.
Type 'q()' to quit R.

> testthat::test_package('testdtbug')
-- 1. Error: dt [ method ok (@test-dt.R#5)  ------------------------------------
object 'foo' not found
1: data.table(group = 1:2, foo = "bar")[, {
       expect_identical(foo, "bar")
   }, by = group] at C:/Users/th798/R/win-library/3.6/testdtbug/tests/testthat/test-dt.R:5
2: `[.data.table`(data.table(group = 1:2, foo = "bar"), , {
       expect_identical(foo, "bar")
   }, by = group) at C:/Users/th798/R/win-library/3.6/testdtbug/tests/testthat/test-dt.R:5
3: `[.data.frame`(x, i, j)
4: expect_identical(foo, "bar") at C:/Users/th798/R/win-library/3.6/testdtbug/tests/testthat/test-dt.R:6
5: quasi_label(enquo(object), label, arg = "object")
6: eval_bare(get_expr(quo), get_env(quo))

== testthat results  ===========================================================
[ OK: 0 | SKIPPED: 0 | WARNINGS: 0 | FAILED: 1 ]
1. Error: dt [ method ok (@test-dt.R#5) 

Error: testthat unit tests failed
Execution halted
]0;MINGW64:/c/Users/th798/R
th798@cmp2986 MINGW64 ~/R
$ R --vanilla -e "testthat::test_file(system.file('tests', 'testthat', 'test-dt.R', package='testdtbug'))"

R version 3.6.1 (2019-07-05) -- "Action of the Toes"
Copyright (C) 2019 The R Foundation for Statistical Computing
Platform: x86_64-w64-mingw32/x64 (64-bit)

R is free software and comes with ABSOLUTELY NO WARRANTY.
You are welcome to redistribute it under certain conditions.
Type 'license()' or 'licence()' for distribution details.

  Natural language support but running in an English locale

R is a collaborative project with many contributors.
Type 'contributors()' for more information and
'citation()' on how to cite R or R packages in publications.

Type 'demo()' for some demos, 'help()' for on-line help, or
'help.start()' for an HTML browser interface to help.
Type 'q()' to quit R.

> testthat::test_file(system.file('tests', 'testthat', 'test-dt.R', package='testdtbug'))
v |  OK F W S | Context
v |   2       | dt

== Results =====================================================================
Duration: 0.1 s

OK:       2
Failed:   0
Warnings: 0
Skipped:  0
> 
> 
]0;MINGW64:/c/Users/th798/R
th798@cmp2986 MINGW64 ~/R
$ R --vanilla -e "devtools::session_info()"

R version 3.6.1 (2019-07-05) -- "Action of the Toes"
Copyright (C) 2019 The R Foundation for Statistical Computing
Platform: x86_64-w64-mingw32/x64 (64-bit)

R is free software and comes with ABSOLUTELY NO WARRANTY.
You are welcome to redistribute it under certain conditions.
Type 'license()' or 'licence()' for distribution details.

  Natural language support but running in an English locale

R is a collaborative project with many contributors.
Type 'contributors()' for more information and
'citation()' on how to cite R or R packages in publications.

Type 'demo()' for some demos, 'help()' for on-line help, or
'help.start()' for an HTML browser interface to help.
Type 'q()' to quit R.

> devtools::session_info()
- Session info ---------------------------------------------------------------
 setting  value                       
 version  R version 3.6.1 (2019-07-05)
 os       Windows 10 x64              
 system   x86_64, mingw32             
 ui       RTerm                       
 language (EN)                        
 collate  English_United States.1252  
 ctype    English_United States.1252  
 tz       America/Phoenix             
 date     2019-09-04                  

- Packages -------------------------------------------------------------------
 package     * version date       lib source           
 assertthat    0.2.1   2019-03-21 [1] CRAN (R 3.6.1)   
 backports     1.1.4   2019-04-10 [1] CRAN (R 3.6.1)   
 callr         3.3.1   2019-07-18 [1] CRAN (R 3.6.1)   
 cli           1.1.0   2019-03-19 [1] CRAN (R 3.6.1)   
 crayon        1.3.4   2017-09-16 [1] CRAN (R 3.6.1)   
 desc          1.2.0   2018-05-01 [1] CRAN (R 3.6.1)   
 devtools      2.1.0   2019-07-06 [1] CRAN (R 3.6.1)   
 digest        0.6.20  2019-07-04 [1] CRAN (R 3.6.1)   
 fs            1.3.1   2019-05-06 [1] CRAN (R 3.6.1)   
 glue          1.3.1   2019-03-12 [1] CRAN (R 3.6.1)   
 magrittr      1.5     2014-11-22 [1] CRAN (R 3.6.1)   
 memoise       1.1.0   2017-04-21 [1] CRAN (R 3.6.1)   
 pkgbuild      1.0.3   2019-03-20 [1] CRAN (R 3.6.1)   
 pkgload       1.0.2   2018-10-29 [1] CRAN (R 3.6.1)   
 prettyunits   1.0.2   2015-07-13 [1] CRAN (R 3.6.1)   
 processx      3.4.1   2019-07-18 [1] CRAN (R 3.6.1)   
 ps            1.3.0   2018-12-21 [1] CRAN (R 3.6.1)   
 R6            2.4.0   2019-02-14 [1] CRAN (R 3.6.1)   
 Rcpp          1.0.2   2019-07-25 [1] CRAN (R 3.6.1)   
 remotes       2.1.0   2019-06-24 [1] standard (@2.1.0)
 rlang         0.4.0   2019-06-25 [1] CRAN (R 3.6.1)   
 rprojroot     1.3-2   2018-01-03 [1] CRAN (R 3.6.1)   
 sessioninfo   1.1.1   2018-11-05 [1] CRAN (R 3.6.1)   
 testthat      2.2.0   2019-07-22 [1] CRAN (R 3.6.1)   
 usethis       1.5.1   2019-07-04 [1] CRAN (R 3.6.1)   
 withr         2.1.2   2018-03-15 [1] CRAN (R 3.6.1)   

[1] C:/Users/th798/R/win-library/3.6
[2] C:/Program Files/R/R-3.6.1/library
> 
> 
]0;MINGW64:/c/Users/th798/R
th798@cmp2986 MINGW64 ~/R
$ 

This is happening both on MacOs (see https://github.com/r-lib/testthat/issues/933) and on windows (as shown above)

[jangorecki]

@tdhock thanks for reporting. Are you able to reproduce the issue when putting NAMESPACE file in your package? I think that NAMESPACE is mandatory for a package to be valid, since 2.14 or 3.0.0.
Did you try .datatable.aware = TRUE described in importing vignette?

[tdhock]

.datatable.aware=TRUE seems to fix the problem, thanks for the suggestion.

May I suggest a change to the importing vignette? It is clear that .datatable.aware is needed if data.table is imported, but it is not clear that it is necessary when it is suggested. (as in my case where I only want to use data.table in vignettes and tests)

It would be great if you could mention that in the section "Optionally import data.table: Suggests" for example

When using a package as a suggested dependency, you should not import it in the NAMESPACE file. Just mention it in the DESCRIPTION file. When using data.table functions in package code (R/* files) you need to use the data.table:: prefix because none of them are imported. When using data.table in package tests (e.g. tests/testthat/test* files), you need to declare .datatable.aware=TRUE in one of the R/* files.

[jangorecki]

@tdhock sure, moreover you are welcome to submit PR directly.

[tdhock]

ok here is the pr https://github.com/Rdatatable/data.table/compare/master...tdhock:patch-1

[tdhock]

i mean here, https://github.com/Rdatatable/data.table/pull/3818