Skip to content

[Bioc-devel] BiocGenerics attaches when methods package is imported

5 messages · Andreas Kapouranis, Joris Meys, Michael Lawrence

Original
Andreas Kapouranis
# 
Hi,

I am developing a package for Bioconductor, and I have a weird issue when
importing the 'methods' package. Before I add the 'methods' package in
Imoprts in the DESCRIPTION file, I pass the CMD check without any warnings.

However, when I add the 'methods' package in the Imports field, I get a
warning during CMD check. (I need the methods package since I use the
methods::is() function, and if I call it just as is(), when running the
package from Rscript will fail since by default it does not load the
'methods' package.)

I read a similar issue here:
https://stat.ethz.ch/pipermail/bioc-devel/2015-October/008193.html
and I updated all my packages from Bioconductor including IRanges and
GenomicRanges.

This is the warning that I get, which is exactly the same as the previous
issue in Bioc-Devel, and I understand that for some reason the BiocGenerics
is attached only when the 'methods' package is imported.

* checking for missing documentation entries ... WARNING

Attaching package: ?BiocGenerics?

The following objects are masked from ?package:parallel?:

    clusterApply, clusterApplyLB, clusterCall, clusterEvalQ,
    clusterExport, clusterMap, parApply, parCapply, parLapply,
    parLapplyLB, parRapply, parSapply, parSapplyLB

The following objects are masked from ?package:stats?:

    IQR, mad, xtabs

The following objects are masked from ?package:base?:

    Filter, Find, Map, Position, Reduce, anyDuplicated, append,
    as.data.frame, as.vector, cbind, colnames, do.call, duplicated,
    eval, evalq, get, grep, grepl, intersect, is.unsorted, lapply,
    lengths, mapply, match, mget, order, paste, pmax, pmax.int, pmin,
    pmin.int, rank, rbind, rownames, sapply, setdiff, sort, table,
    tapply, union, unique, unlist, unsplit

All user-level objects in a package should have documentation entries.
See chapter ?Writing R documentation files? in the ?Writing R
Extensions? manual.



My DESCRIPTION file when I add the 'methods' package is the following:
...
Depends: R (>= 3.2.0)
LazyData: TRUE
RoxygenNote: 5.0.1
Imports: GenomicRanges,
    IRanges,
    assertthat,
    S4Vectors,
    graphics,
    data.table,
    methods
Suggests: knitr,
    rmarkdown,
    testthat
VignetteBuilder: knitr



Regards,
Andreas
Joris Meys
# 
Hi Andreas,

for a good understanding: just mentioning the methods package in the
DESCRIPTION file isn't doing a thing if you don't also add

import(methods)

to the NAMESPACE file.

So it looks like the warning occurs when the package indicates that the
methods package is imported when it actually isn't. Do you still get that
warning if you update the NAMESPACE file?

Cheers
Joris

On Mon, Feb 15, 2016 at 4:52 PM, Andreas Kapouranis <
kapouranis.andreas at gmail.com> wrote:

            

      
      
      
    

        

      
      
    

        

  
    

      
      
    

  


  


      

    

    

      
      

        


  

        

      Michael Lawrence
    

    

      
      #
    

  


  

      
It would be helpful to have a reproducible example, i.e., your actual
package or a skeleton of it that reproduces the issue. Btw, you might
be able to debug this behavior by:

debug(tools::undoc)
tools::undoc(dir=yourPackageDir)



On Mon, Feb 15, 2016 at 7:52 AM, Andreas Kapouranis

        
<kapouranis.andreas at gmail.com> wrote:

            

      
      
      
    

        

      
      
    

  
  


  


      

    

    

      
      

        


  

        

      Andreas Kapouranis
    

    

      
      #
    

  


  

      
Hi Joris,

Apologies for not including the NAMESPACE file, but I have included the
methods package also there, although I have used:

importFrom(methods,is)

since I just need the is() function for the time being.

Regards,
Andreas

----------------------------------------------------------
Chantriolnt - Andreas Kapourani
PhD Candidate in Data Science,
School of Informatics,
University of Edinburgh

e-mail  : c.a.kapourani at sms.ed.ac.uk  <c.a.kapourani at sms.ed.ac.uk>
           : kapouranis.andreas at gmail.com

        
On Mon, Feb 15, 2016 at 4:16 PM, Joris Meys <Joris.Meys at ugent.be> wrote:

        

            

      
      
      
    

        

      
      
    

            

      
      
      
    

        

  
  


  


      

    

    

      
      

        


  

        

      Andreas Kapouranis
    

    

      
      #
    

  


  

      
Hi Michael,

Thanks for the reply.

I created a minimal reproducible package with the issue. As it seems the
warning comes because I have a 'data' folder where I have stored some
example data, and one of the data is stored in GRanges format.
If I delete the GRanges example data and keep any other data, I get no
warning.

So the issue is only when I import the 'methods' package and I have data
locally stored as GRanges objects in the 'data' folder.

You can download and install the simple example package by typing:

devtools::install_github("andreaskapou/simplePackage")


Or you can see the code in the following link, which just contains a simple
function to call the methods::is() function and create a GRanges object.
https://github.com/andreaskapou/simplePackage

Thanks,
Andreas

----------------------------------------------------------
Chantriolnt - Andreas Kapourani
PhD Candidate in Data Science,
School of Informatics,
University of Edinburgh

e-mail  : c.a.kapourani at sms.ed.ac.uk  <c.a.kapourani at sms.ed.ac.uk>
           : kapouranis.andreas at gmail.com

On Mon, Feb 15, 2016 at 4:22 PM, Michael Lawrence <lawrence.michael at gene.com