R CMD check and Rcpp.package.skeleton()

[coatless]

Per Creating R package with Rcpp: errors, it was noted that the CRAN check fails capriciously on the description file along with: documentation failure and dependency check issues.

Using Rcpp.package.skeleton() under the current CRAN build, Rcpp has the following CRAN check errors:

0 errors | 3 warnings | 2 notes
checking whether package ‘an2’ can be installed ... WARNING

checking Rd files ... WARNING
prepare_Rd: an2-package.Rd:26: All text must be in a section
prepare_Rd: an2-package.Rd:32-34: Dropping empty section \examples

checking Rd cross-references ... WARNING
Unknown package ‘<pkg>’ in Rd xrefs

checking DESCRIPTION meta-information ... NOTE
Malformed Description field: should contain one or more complete sentences.

checking dependencies in R code ... NOTE
Namespace in Imports field not imported from: ‘Rcpp’
  All declared Imports should be used.

Dirk already set up a patch for the description check update-pkg-skeleton (see PR #471) leading to new Rcpp skeleton packages having:

checking Rd files ... WARNING
prepare_Rd: anRpackage-package.Rd:26: All text must be in a section
prepare_Rd: anRpackage-package.Rd:32-34: Dropping empty section \examples

checking Rd cross-references ... WARNING
Unknown package ‘<pkg>’ in Rd xrefs

checking dependencies in R code ... NOTE
Namespace in Imports field not imported from: ‘Rcpp’
  All declared Imports should be used.

I would highly suggest the addition of importFrom(Rcpp,evalCpp) to the NAMESPACE file so that users are only forced to modify one file the *-package.Rd.

[eddelbuettel]

As I mentioned in the StackOverlow comment of the thread where this started, the issue

checking Rd files ... WARNING
prepare_Rd: an2-package.Rd:26: All text must be in a section
prepare_Rd: an2-package.Rd:32-34: Dropping empty section \examples

has been taking care of for years (!!) by farming out to pkgKitten which has a Suggests: from Rcpp itself. So I consider this issue closed. You can try to convince me otherwise.

Now, before I committed 47df44a and created the PRed that is now in #471 I also tested the thusly-created package... and got it again to be squeaky-clean (provided you have pkgKitten) under R-release and R-devel.

The whole issue is somewhat ridiculous as we have package creators in

• R itself (but sucky, hence pkgKitten)
• pkgKitten
• RStudio
• devtools (I think)

and probably a few more places I do not know about.

[eddelbuettel]

Below is a full log using the Rcpp code in the current PR. I am inclined to close this. Thoughts?

edd@max:/tmp$ r -e'Rcpp::Rcpp.package.skeleton("coatlessDemo")'
Creating directories ...
Creating DESCRIPTION ...
Creating NAMESPACE ...
Creating Read-and-delete-me ...
Saving functions and data ...
Making help files ...
Done.
Further steps are described in './coatlessDemo/Read-and-delete-me'.

Adding Rcpp settings
 >> added Imports: Rcpp
 >> added LinkingTo: Rcpp
 >> added useDynLib directive to NAMESPACE
 >> added importFrom(Rcpp, evalCpp) directive to NAMESPACE
 >> added example src file using Rcpp attributes
 >> compiled Rcpp attributes
 >> added Rd file for rcpp_hello_world
edd@max:/tmp$ R CMD build coatlessDemo
* checking for file ‘coatlessDemo/DESCRIPTION’ ... OK
* preparing ‘coatlessDemo’:
* checking DESCRIPTION meta-information ... OK
* cleaning src
* checking for LF line-endings in source and make files
* checking for empty or unneeded directories
* building ‘coatlessDemo_1.0.tar.gz’

edd@max:/tmp$ RD --version
R Under development (unstable) (2016-05-07 r70591) -- "Unsuffered Consequences"
Copyright (C) 2016 The R Foundation for Statistical Computing
Platform: x86_64-pc-linux-gnu (64-bit)

R is free software and comes with ABSOLUTELY NO WARRANTY.
You are welcome to redistribute it under the terms of the
GNU General Public License versions 2 or 3.
For more information about these matters see
http://www.gnu.org/licenses/.

edd@max:/tmp$ RD CMD check --as-cran coatlessDemo_1.0.tar.gz 
* using log directory ‘/tmp/coatlessDemo.Rcheck’
* using R Under development (unstable) (2016-05-07 r70591)
* using platform: x86_64-pc-linux-gnu (64-bit)
* using session charset: UTF-8
* using option ‘--as-cran’
* checking for file ‘coatlessDemo/DESCRIPTION’ ... OK
* checking extension type ... Package
* this is package ‘coatlessDemo’ version ‘1.0’
* checking CRAN incoming feasibility ... NOTE
Maintainer: ‘Your Name <your@email.com>’

New submission
* checking package namespace information ... OK
* checking package dependencies ... OK
* checking if this is a source package ... OK
* checking if there is a namespace ... OK
* checking for executable files ... OK
* checking for hidden files and directories ... OK
* checking for portable file names ... OK
* checking for sufficient/correct file permissions ... OK
* checking whether package ‘coatlessDemo’ can be installed ... OK
* checking installed package size ... OK
* checking package directory ... OK
* checking DESCRIPTION meta-information ... OK
* checking top-level files ... OK
* checking for left-over files ... OK
* checking index information ... OK
* checking package subdirectories ... OK
* checking R files for non-ASCII characters ... OK
* checking R files for syntax errors ... OK
* checking whether the package can be loaded ... OK
* checking whether the package can be loaded with stated dependencies ... OK
* checking whether the package can be unloaded cleanly ... OK
* checking whether the namespace can be loaded with stated dependencies ... OK
* checking whether the namespace can be unloaded cleanly ... OK
* checking loading without being on the library search path ... OK
* checking use of S3 registration ... OK
* checking dependencies in R code ... OK
* checking S3 generic/method consistency ... OK
* checking replacement functions ... OK
* checking foreign function calls ... OK
* checking R code for possible problems ... OK
* checking Rd files ... OK
* checking Rd metadata ... OK
* checking Rd line widths ... OK
* checking Rd cross-references ... OK
* checking for missing documentation entries ... OK
* checking for code/documentation mismatches ... OK
* checking Rd \usage sections ... OK
* checking Rd contents ... OK
* checking for unstated dependencies in examples ... OK
* checking line endings in C/C++/Fortran sources/headers ... OK
* checking compiled code ... OK
* checking examples ... OK
* checking PDF version of manual ... OK
* DONE

Status: 1 NOTE
See
  ‘/tmp/coatlessDemo.Rcheck/00check.log’
for details.


edd@max:/tmp$ 

[coatless]

Here it goes...

I do not think the install base has pkgKitten or even the understanding behind pkgKitten is known to the larger community. As the name of pkgKitten does not convey its intent outside of the internet's affinity for cats. So, while it might solve this issue entirely, as pointed out above, it does so for a small percentage of the community that could likely diagnosis and fix the issue themselves.

See:
pkgKitten tracked DL: vs. Rcpp tracked DL:

Under current rates that is: 22e3/4e+06: 0.0055

(Note: The above number argument has flaws as r-pkg mirror does not differentiate between versions nor does it contain all the data as it is only provided by RStudio's CDN)

With this being said, it is particularly problematic as users who are creating skeleton packages have little experience or are being told this is the correct way to embed c++ code under the Rcpp scheme. Furthermore, the importFrom() error and the lack of a useDynLib() statement within NAMESPACE has been shown to cause issues that frequently end up on SO.

Thus, I would suggest changing the dependence with pkgKitten so it is automatically included or directly fixing the generation within the the skeleton version of Rcpp.

Apologies for the delay as I am now traveling

[eddelbuettel]

The download count does not matter all that much because many people just run code with Rcpp--thing dplyr and all its friends. And for those actually writing code with Rcpp, I am thinking that a Suggests: should do. Just as it does for RUnit, say.

One thing we could of course do is to add a note in the Rcpp FAQ. pkgKitten is not yet mentioned in any vignettes.

In any event, I hear you---I am as bugged as you are by packages created freshly by proper tools (like package.skeleton()) yet which fail to cleanly pass R CMD check. But I want to treat lightly on additional Depends:. It really is just for one function which we could even duplicate (as inelegant as that is).

Any thought from @jjallaire @kevinushey @thirdwing ?

[jjallaire]

I'd be in favor of duplicating it so that we are always guaranteed to
create a clean package.

On Tue, May 10, 2016 at 6:21 PM, Dirk Eddelbuettel <notifications@github.com

wrote:

The download count does not matter all that much because many people just
run code with Rcpp--thing dplyr and all its friends. And for those
actually writing code with Rcpp, I am thinking that a Suggests: should do.
Just as it does for RUnit, say.

One thing we could of course do is to add a note in the Rcpp FAQ.
pkgKitten is not yet mentioned in any vignettes.

In any event, I hear you---I am as bugged as you are by packages created
freshly by proper tools (like package.skeleton()) yet which fail to
cleanly pass R CMD check. But I want to treat lightly on additional
Depends:. It really is just for one function which we could even duplicate
(as inelegant as that is).

Any thought from @jjallaire https://github.com/jjallaire @kevinushey
https://github.com/kevinushey @thirdwing https://github.com/thirdwing
?

—
You are receiving this because you were mentioned.
Reply to this email directly or view it on GitHub
#472 (comment)

[eddelbuettel]

I think I agree.

We can keep the 'Suggests: pkgKitten' as a free option in case functionality on that side gets added. But otherwise, it really is just these few lines so we may as well import them. I'll get to that tomorrow.

[eddelbuettel]

@coatless I made one more change and this is now in the mainline. Can you take a peek? Should now create a purring little package even when pkgKitten is not around to help.

[coatless]

@eddelbuettel

d635089 's Rcpp.package.skeleton() generated package passes the CRAN check as tested on Windows 10 with R 3.3.0.

Thanks!

(Misc Note: It may yield issues under devtools::check() since roxygen2 is not used to generate the NAMESPACE. But, that is minor and not very important as devtools::check(document = F) fixes it plus the console text indicates the different package components for Rcpp.)

[eddelbuettel]

@coatless Thanks.

Re your misc note: Whether or not you let roxygen2 play with DESCRIPTION and/or NAMESPACE should be entirely orthogonal to this package. The only relevant commands, really, are R CMD ... as in WRE.