Update package to pass R CMD CHECK for CRAN submission

[ramhiser]

The package was removed from CRAN because there's now a warning.

checking for unstated dependencies in ‘tests’ ... WARNING
'library' or 'require' call not declared from: ‘testthat’

According to this testthat issue, need to add Suggests: testthat to the description file.

[ramhiser]

The check passes on Mac (local) and Ubuntu (on Travis-CI).

The Windows build is failing with 1 note, 1 warning, and 1 error:

Note

* checking CRAN incoming feasibility ... NOTE
Maintainer: 'John A. Ramey <johnramey@gmail.com>'

New submission

Package was archived on CRAN

CRAN repository db overrides:
  X-CRAN-Comment: Archived on 2017-04-21 as check errors were not
    corrected despite reminders.

Found the following (possibly) invalid URLs:
  URL: http://cran.r-project.org/package=sparsediscrim
    From: README.md
    CRAN URL not in canonical form
  Canonical CRAN.R-project.org URLs use https.

Warning

* checking whether package 'sparsediscrim' can be installed ... WARNING
Found the following significant warnings:
  Warning: Installed Rcpp (0.12.11) different from Rcpp used to build dplyr (0.12.10).
  Warning: Installed R (R Under development (unstable) (2017-05-20 r72713)) different from R used to build dplyr (R Under development (unstable) (2017-05-20 r72708)).
See 'd:/RCompile/CRANguest/R-devel/sparsediscrim.Rcheck/00install.out' for details.

Error

** running tests for arch 'i386' ... [10s] ERROR
  Running 'testthat.r' [9s]
Running the tests in 'tests/testthat.r' failed.
Last 13 lines of output:
  [3]  -0.00192 -  0.00192 == -0.00385
  [4]   0.00376 - -0.00376 ==  0.00753
  [5]  -0.07789 -  0.07789 == -0.15578
  [9]  -0.00192 -  0.00192 == -0.00385
  [13]  0.00376 - -0.00376 ==  0.00753
  
  
  testthat results ================================================================
  OK: 160 SKIPPED: 0 FAILED: 3
  1. Failure: HDRDA's calculations are correct for (lambda, gamma) = (0, 0) (@test-hdrda.r#60) 
  2. Failure: HDRDA's calculations are correct for (lambda, gamma) = (0, 0) (@test-hdrda.r#61) 
  3. Failure: HDRDA's calculations are correct for (lambda, gamma) = (0, 0) (@test-hdrda.r#62) 
  
  Error: testthat unit tests failed
  Execution halted

[DarioS]

Will this happen soon? I have a Bioconductor package that depends on sparsediscrim that is no longer able to be built and needs to be resolved before the next version of the repository is released.

[ramhiser]

@DarioS Yes! I'm blocking out some time this weekend. Sorry about the inconvenience.

[DarioS]

Or this weekend?

[ramhiser]

Running on latest R version 3.4.1. devtools::check() everything passes.

Using devtools::build_win(), tests pass on x64 on Windows, but not on i386.

** running tests for arch 'i386' ... [12s] ERROR
  Running 'testthat.r' [10s]
Running the tests in 'tests/testthat.r' failed.
Last 13 lines of output:
  [3]  -0.00192 -  0.00192 == -0.00385
  [4]   0.00376 - -0.00376 ==  0.00753
  [5]  -0.07789 -  0.07789 == -0.15578
  [9]  -0.00192 -  0.00192 == -0.00385
  [13]  0.00376 - -0.00376 ==  0.00753
  
  
  testthat results ================================================================
  OK: 160 SKIPPED: 0 FAILED: 3
  1. Failure: HDRDA's calculations are correct for (lambda, gamma) = (0, 0) (@test-hdrda.r#60) 
  2. Failure: HDRDA's calculations are correct for (lambda, gamma) = (0, 0) (@test-hdrda.r#61) 
  3. Failure: HDRDA's calculations are correct for (lambda, gamma) = (0, 0) (@test-hdrda.r#62) 
  
  Error: testthat unit tests failed
  Execution halted
** running tests for arch 'x64' ... [11s] OK
  Running 'testthat.r' [11s]

[ramhiser]

Submitted updated package. We'll see if it makes it through the CRAN gauntlet.

[ramhiser]

That Windows i386 test failure above came back to haunt me. CRAN denied package submission. Think I'm gonna remove that test for now. I don't have a Windows box to investigate. Especially difficult since it passes on Windows x64.

[ramhiser]

I submitted another version 2 hours ago. It hasn't been rejected yet, so I think it made it through CRAN's automated checks.

Hopefully, it makes it through the manual BDR check tomorrow. :-\

[DarioS]

I have a i386 Windows computer. It turns out that the sign of some of the off-diagonal numbers are the opposite to what is expected.

Browse[1]> W_k$setosa
             [,1]        [,2]         [,3]        [,4]
[1,]  0.158807209  0.08873751 -0.033947619 0.008166478
[2,]  0.088737511  0.09779328 -0.024209703 0.013101441
[3,] -0.033947619 -0.02420970  0.032950803 0.003264817
[4,]  0.008166478  0.01310144  0.003264817 0.013468712
Browse[1]> W_k_expected$setosa
             [,1]        [,2]         [,3]         [,4]
[1,]  0.158807209 -0.08873751  0.033947619 -0.008166478
[2,] -0.088737511  0.09779328 -0.024209703  0.013101441
[3,]  0.033947619 -0.02420970  0.032950803  0.003264817
[4,] -0.008166478  0.01310144  0.003264817  0.013468712

It's likely that the eigenvectors used in the calculation of W_k have the opposite sign on a 32-bit computer.

[ramhiser]

Just got the note.

Thanks, on CRAN now.

For your next version:
Can you provide some references in the 'Description' field of your DESCRIPTION file in the form
authors (year) <doi:...> or <arXiv:...>
(no space after 'doi:' and 'arXiv:')?