[R-pkg-devel] Check Results ERROR on Windows R release

2019-05-17 Thread Jarrett Phillips 
Hello,

On submitting an update of my R package (HACSim), checks fail on
r-release-windows-ix86+x86_64


The updated package, in addition to improving on documentation, also fixes
a noLD error by replacing

if (sum(probs) == 1) {
   stop("probs must sum to 1")
}

with the less strict

if (!isTRUE(all.equal(1, sum(probs), tolerance =
.Machine$double.eps^0.25))) {
  stop("probs must sum to 1")
}

On viewing the check details, the following was noted:


** byte-compile and prepare package for lazy loading

Error in loadNamespace(j <- i[[1L]], c(lib.loc, .libPaths()),
versionCheck = vI[[j]]) :
  there is no package called 'dplyr'
Calls:  ... loadNamespace -> withRestarts -> withOneRestart
-> doWithOneRestart
Execution halted
ERROR: lazy loading failed for package 'HACSim'


I don't know why the dplyr package creeps in, as I don't list it in my
NAMESPACE file.

Any idea on what could be going on here?


Thanks.

Cheers,

Jarrett

[[alternative HTML version deleted]]

__
R-package-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-package-devel

Re: [R-pkg-devel] Check Results ERROR on Windows R release

2019-05-17 Thread Winston Chang 
It looks like hacSIM imports pegas, which imports adegenet, which imports dplyr.

-Winston


On Fri, May 17, 2019 at 11:04 AM Jarrett Phillips
 wrote:
>
> Hello,
>
> On submitting an update of my R package (HACSim), checks fail on
> r-release-windows-ix86+x86_64
> 
>
> The updated package, in addition to improving on documentation, also fixes
> a noLD error by replacing
>
> if (sum(probs) == 1) {
>stop("probs must sum to 1")
> }
>
> with the less strict
>
> if (!isTRUE(all.equal(1, sum(probs), tolerance =
> .Machine$double.eps^0.25))) {
>   stop("probs must sum to 1")
> }
>
> On viewing the check details, the following was noted:
>
>
> ** byte-compile and prepare package for lazy loading
>
> Error in loadNamespace(j <- i[[1L]], c(lib.loc, .libPaths()),
> versionCheck = vI[[j]]) :
>   there is no package called 'dplyr'
> Calls:  ... loadNamespace -> withRestarts -> withOneRestart
> -> doWithOneRestart
> Execution halted
> ERROR: lazy loading failed for package 'HACSim'
>
>
> I don't know why the dplyr package creeps in, as I don't list it in my
> NAMESPACE file.
>
> Any idea on what could be going on here?
>
>
> Thanks.
>
> Cheers,
>
> Jarrett
>
> [[alternative HTML version deleted]]
>
> __
> R-package-devel@r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-package-devel

__
R-package-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-package-devel

Re: [R-pkg-devel] Check Results ERROR on Windows R release

2019-05-17 Thread Iñaki Ucar 
On Fri, 17 May 2019 at 18:05, Jarrett Phillips
 wrote:
>
> Hello,
>
> On submitting an update of my R package (HACSim), checks fail on
> r-release-windows-ix86+x86_64
> 
>
> The updated package, in addition to improving on documentation, also fixes
> a noLD error by replacing
>
> if (sum(probs) == 1) {
>stop("probs must sum to 1")
> }
>
> with the less strict
>
> if (!isTRUE(all.equal(1, sum(probs), tolerance =
> .Machine$double.eps^0.25))) {
>   stop("probs must sum to 1")
> }
>
> On viewing the check details, the following was noted:
>
>
> ** byte-compile and prepare package for lazy loading
>
> Error in loadNamespace(j <- i[[1L]], c(lib.loc, .libPaths()),
> versionCheck = vI[[j]]) :
>   there is no package called 'dplyr'
> Calls:  ... loadNamespace -> withRestarts -> withOneRestart
> -> doWithOneRestart
> Execution halted
> ERROR: lazy loading failed for package 'HACSim'
>
>
> I don't know why the dplyr package creeps in, as I don't list it in my
> NAMESPACE file.

Your package imports pegas, which depends on adegenet, which imports dplyr.

> Any idea on what could be going on here?

Probably just a glitch that will be solved the next time the check is triggered.

Iñaki

__
R-package-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-package-devel

Re: [R-pkg-devel] R CMD check works but with devtools::check() examples fail

2019-05-17 Thread Dirk Eddelbuettel 


On 17 May 2019 at 08:33, Rainer Krug wrote:
| Interesting discussion - because I had similar issues.
| 
| @Dirk: You are developing packages fir ages - do you probably have a template 
make file you could share (as I assume, you use make) which is doing all these 
checks?

Nothing special -- just a few simple wrappers (generally written using
littler's r as I tend to live on Linux, should work from Rscript too) which I
call from shell: build.r to build, check.r to check, rcc.r to run (Gabor's
nice) rcmdcheck around the tarball (or directory or ...), install.r and
install2.r, update.r and so on. They are in the inst/examples/ directory of
source littler (and hence examples/ of installed littler).

The rule is that they are mostly skinny wrapper around the same functions R
uses when your call R CMD build or R CMD check, but saving me a few
keystrokes each time ... and cooperating better with TAB completion in bash.

So nothing magic.  Sorry :)

Dirk

PS It would be nice if R Core exported these functions:
 tools:::.build_packages
 tools:::.check_packages
   for us to use from packages.

-- 
http://dirk.eddelbuettel.com | @eddelbuettel | e...@debian.org

__
R-package-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-package-devel

Re: [R-pkg-devel] Check Results ERROR on Windows R release

2019-05-17 Thread Uwe Ligges 

Yes, looks like some race condition, just wait a day or two.

Best,
Uwe Ligges

On 17.05.2019 18:52, Iñaki Ucar wrote:

On Fri, 17 May 2019 at 18:05, Jarrett Phillips
 wrote:


Hello,

On submitting an update of my R package (HACSim), checks fail on
r-release-windows-ix86+x86_64


The updated package, in addition to improving on documentation, also fixes
a noLD error by replacing

if (sum(probs) == 1) {
stop("probs must sum to 1")
}

with the less strict

if (!isTRUE(all.equal(1, sum(probs), tolerance =
.Machine$double.eps^0.25))) {
   stop("probs must sum to 1")
}

On viewing the check details, the following was noted:


** byte-compile and prepare package for lazy loading

Error in loadNamespace(j <- i[[1L]], c(lib.loc, .libPaths()),
versionCheck = vI[[j]]) :
   there is no package called 'dplyr'
Calls:  ... loadNamespace -> withRestarts -> withOneRestart
-> doWithOneRestart
Execution halted
ERROR: lazy loading failed for package 'HACSim'


I don't know why the dplyr package creeps in, as I don't list it in my
NAMESPACE file.


Your package imports pegas, which depends on adegenet, which imports dplyr.


Any idea on what could be going on here?


Probably just a glitch that will be solved the next time the check is triggered.

Iñaki

__
R-package-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-package-devel



__
R-package-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-package-devel