[Bioc-devel] class union with AffyBatch

Martin Morgan mtmorgan at fhcrc.org
Thu Oct 9 19:40:47 CEST 2008


Hi Audrey --

audrey <audrey at ebi.ac.uk> writes:

> Hi all,
>
> In my package, I am trying to set a class union the following way:
> setClassUnion("OneColour", c("ExpressionSet", "AffyBatch"))
>
> In the NAMESPACE if I write the following:
> importClassesFrom("Biobase", "ExpressionSet")
> importClassesFrom("Biobase", "AffyBatch")

definitely the last line is not correct.

> When I build the package I have this error message
> Error: package 'affy' does not have a name space
> Execution halted
>
> If I remove the importClassesFrom("Biobase", "AffyBatch") from my
> NAMESPACE, I have this warning:
> Warning in .findOrCopyClass(class1, classDef, where, "superClass") :
>   Class "AffyBatch" is defined (with package slot "affy") but no
> metadata object found to revise superClass information---not exported?
> Making a copy in package "arrayQualityMetrics"
>
> In the dependencies in the DESCRIPTION, I do depend on the package
> "affy". This is using R2.8 and affy_1.19.0.
>
> Any idea on how I could solve that? Is it possible to import a class
> from a package that does not have a namespace?

I think there's something else going on. If I create a package with

DESCRIPTION
-----------
Depends: affy
Imports: Biobase, methods

NAMESPACE
---------
import(methods)
importClassesFrom(Biobase, "ExpressionSet")
exportClasses("OneColour")

R/tmp.R
-------
setClassUnion("OneColour", c("ExpressionSet", "AffyBatch"))

I can build and check without error with

> R --version
R version 2.8.0 beta (2008-10-09 r46680)

In arrayQualityMetrics, if I add a file like tmp.R, move methods to
Imports: (thinking that the end user doesn't need access to the
functionality of the methods package to use my package), add
imports(methods) and exportClasses("OneColour") then again I can build
and check without errors / warnings.

My only real suggestion is to update R.

Sorry to be less than helpful

Martin

> Thanks
> Audrey
>
> -- 
> Audrey Kauffmann
> EMBL - EBI
> Cambridge UK
> http://www.ebi.ac.uk/~audrey
>
> _______________________________________________
> Bioc-devel at stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/bioc-devel

-- 
Martin Morgan
Computational Biology / Fred Hutchinson Cancer Research Center
1100 Fairview Ave. N.
PO Box 19024 Seattle, WA 98109

Location: Arnold Building M2 B169
Phone: (206) 667-2793



More information about the Bioc-devel mailing list