Skip to content

[Bioc-devel] Help with R CMD check NOTEs

5 messages · Anusha Nagari, Hervé Pagès, Pariksheet Nanda

#
Hello,

I am getting the following NOTES at R CMD check step:

NOTE1:
* checking package dependencies ... NOTE
Depends: includes the non-default packages:
  ?MASS? ?parallel? ?S4Vectors? ?IRanges? ?GenomeInfoDb?
  ?GenomicRanges? ?GenomicAlignments? ?rtracklayer?
Adding so many packages to the search path is excessive and importing
selectively is preferable.

NOTE2:
* checking re-building of vignette outputs ... NOTE
Warnings in re-building vignettes:
  Warning: file stem ?/fig2? is not portable
  Warning: file stem ?/fig3? is not portable

Can anyone suggest how I can fix these? Thanks in advance!

Best,
Anusha

________________________________

UT Southwestern


Medical Center



The future of medicine, today.
#
Hi Anusha

On Wed, Oct 18, 2017 at 12:04 PM, Anusha Nagari
<Anusha.Nagari at utsouthwestern.edu> wrote:
Move those to the "Imports" section in your package DESCRIPTION file.
Hmm... I think we'll have to look at the exact vignette to see what's
going on.  Presumably that's a LaTeX vignette.  Can you advise the
package name you are working on and/or link to the the source code?
Pariksheet

---
Pariksheet Nanda
PhD Candidate in Genetics and Genomics
System Administrator, Storrs HPC Cluster
University of Connecticut
#
Hi,
On 10/18/2017 09:19 AM, Pariksheet Nanda wrote:
It's important to realize that this will remove all the functionalities
provided by these packages from the search path e.g. your users will
need to explicitly library(MASS) or library(parallel) etc... before
they can access things defined in these packages. The code in your
examples (man page and vignettes) and possibly in your unit tests will
also need to be modified to reflect this.

In other words putting everything in Imports can hurt
usability/friendliness. The approach should be more nuanced. It's
good that the 'R CMD check' NOTE reminds us about the Depends vs
Imports trade-off but for Bioconductor packages the choices made
by the developers will not always please 'R CMD check'.

Cheers,
H.

  
    
#
Hi Herve,

Thank you for your response.
I will leave them as dependencies in the Description file.

Can you please let me know how to go about the following NOTE. Or if this is something that should be really taken care of for a successful package build and install:
* checking re-building of vignette outputs ... NOTE
Warnings in re-building vignettes:
  Warning: file stem ?/fig2? is not portable
  Warning: file stem ?/fig3? is not portable

@Pariksheet: I am working on the groHMM package. https://github.com/Kraus-Lab/groHMM

Thank you,
Anusha
--------------------
From: Anusha Nagari <Anusha.Nagari at UTSouthwestern.edu>

Hi,
On 10/18/2017 09:19 AM, Pariksheet Nanda wrote:
It's important to realize that this will remove all the functionalities
provided by these packages from the search path e.g. your users will
need to explicitly library(MASS) or library(parallel) etc... before
they can access things defined in these packages. The code in your
examples (man page and vignettes) and possibly in your unit tests will
also need to be modified to reflect this.

In other words putting everything in Imports can hurt
usability/friendliness. The approach should be more nuanced. It's
good that the 'R CMD check' NOTE reminds us about the Depends vs
Imports trade-off but for Bioconductor packages the choices made
by the developers will not always please 'R CMD check'.

Cheers,
H.
--
Herv? Pag?s

Program in Computational Biology
Division of Public Health Sciences
Fred Hutchinson Cancer Research Center
1100 Fairview Ave. N, M1-B514
P.O. Box 19024
Seattle, WA 98109-1024

E-mail: hpages at fredhutch.org<https://stat.ethz.ch/mailman/listinfo/bioc-devel>
Phone:  (206) 667-5791
Fax:    (206) 667-1319


From: Anusha Nagari <Anusha.Nagari at UTSouthwestern.edu>
Date: Wednesday, October 18, 2017 at 11:04 AM
To: "bioc-devel at r-project.org" <bioc-devel at r-project.org>
Subject: Help with R CMD check NOTEs

Hello,

I am getting the following NOTES at R CMD check step:

NOTE1:
* checking package dependencies ... NOTE
Depends: includes the non-default packages:
  ?MASS? ?parallel? ?S4Vectors? ?IRanges? ?GenomeInfoDb?
  ?GenomicRanges? ?GenomicAlignments? ?rtracklayer?
Adding so many packages to the search path is excessive and importing
selectively is preferable.

NOTE2:
* checking re-building of vignette outputs ... NOTE
Warnings in re-building vignettes:
  Warning: file stem ?/fig2? is not portable
  Warning: file stem ?/fig3? is not portable

Can anyone suggest how I can fix these? Thanks in advance!

Best,
Anusha

________________________________

UT Southwestern


Medical Center



The future of medicine, today.
#
Hi Anusha,

On Wed, Oct 18, 2017 at 2:30 PM, Anusha Nagari <
Anusha.Nagari at utsouthwestern.edu> wrote:
is something that should be really taken care of for a successful package
build and install:
https://github.com/Kraus-Lab/groHMM

My best guess is you might want to revise your figure label names without
the number.  LaTeX commands like macro names consider numbers to be an
invalid character class.  So you could try replacing instances of fig2 and
fig3 in the vignette with something like figTwo and figThree.

I couldn't reproduce the build NOTE to confirm the fault / fix.  The
bioconductor.org 1.10.0 tarball doesn't see to produce the error, BiocCheck
fails, and I had trouble building the vignette from the GitHub master
branch.

Hope that helps.
Pariksheet