[Bioc-devel] BioC 3.7 Windows check warning "file link zz in package yy does not exist "

Shepherd, Lori Lori.Shepherd at RoswellPark.org
Thu Apr 26 13:13:55 CEST 2018

The packages will not be rejected for this WARNING, but please see the suggestion in the previous response that also might help clear them up.


Lori Shepherd

Bioconductor Core Team

Roswell Park Cancer Institute

Department of Biostatistics & Bioinformatics

Elm & Carlton Streets

Buffalo, New York 14263

From: Bioc-devel <bioc-devel-bounces at r-project.org> on behalf of Joris Meys <Joris.Meys at ugent.be>
Sent: Thursday, April 26, 2018 4:30:08 AM
To: Rahmatallah, Yasir
Cc: bioc-devel
Subject: Re: [Bioc-devel] BioC 3.7 Windows check warning "file link zz in package yy does not exist "

On Wed, Apr 25, 2018 at 11:52 PM, Rahmatallah, Yasir <YRahmatallah at uams.edu>

> Dear all,
> Build report show that many packages still have check warnings due to
> missing file links (cross referencing pages from other packages) under
> windows. The issue happens because there are different man pages for
> different operating systems. My package (GSAR) shows the same warning under
> Windows in the build report, although it locally passes R CMD build and R
> CMD check without warnings. Although not a perfect solution, substituting
> the linking in the .Rd file \code{\link[pkg:bar]{foo}} with
> \code{\link{foo}} or \code{\link[pkg]{foo}} was suggested in a recent
> discussion
> https://stat.ethz.ch/pipermail/bioc-devel/2018-April/013315.html

Actually, that is the correct solution. As per "Writing R Extensions",
\link[pkg]{name} searches for a html file with the name name.html in
package pkg. In a number of cases the name of the html help file is not the
same as the function you want to point to. For example, ppois and qpois etc
are discussed in a page called Poisson. Also in "Writing R Extensions" they
mention that if you want to use another name as a link, you should do
\link[pkg:name]{foo}. In our example: \link[stats:Poisson]{ppois}


> I understood that although a check warning is produced (under Windows
> only), it will be accepted (considered as advice) and packages that
> currently show it will proceed to Bioc release. Is that correct? There are
> still many packages showing this warning in yesterday's report
> http://bioconductor.org/checkResults/3.7/bioc-LATEST/#show=warnings

As this is the case for even recommended packages like the survival
package, I don't see how this would be a reason to reject a package. But it
still makes sense to adapt your links in the way it is specified in the
official R manual.


Joris Meys
Statistical consultant

Department of Data Analysis and Mathematical Modelling
Ghent University
Coupure Links 653, B-9000 Gent (Belgium)

tel: +32 (0)9 264 61 79
Biowiskundedagen 2017-2018

Disclaimer : http://helpdesk.ugent.be/e-maildisclaimer.php

        [[alternative HTML version deleted]]

Bioc-devel at r-project.org mailing list

This email message may contain legally privileged and/or confidential information.  If you are not the intended recipient(s), or the employee or agent responsible for the delivery of this message to the intended recipient(s), you are hereby notified that any disclosure, copying, distribution, or use of this email message is prohibited.  If you have received this message in error, please notify the sender immediately by e-mail and delete this email message from your computer. Thank you.
	[[alternative HTML version deleted]]

More information about the Bioc-devel mailing list