[Bioc-devel] [EXT] Re: Sudden Bioconductor nightly build fail for IsoCorrectoRGUI: cannnot add binding of 'variable' to the base environment

This is because of a change in base R

   https://github.com/wch/r-source/blob/9ad27a744572fca453665e97110ded41a2680a57/doc/NEWS.Rd#L257

It can be reproduced by using a recent version R-4-1.

Looking at your code, I guess it is at

   https://code.bioconductor.org/browse/IsoCorrectoRGUI/blob/master/R/IsoCorrectionGUI.R#L30

that things start to run into trouble -- probably instead you want to create a 'package local' environment, e.g.,

    isoCorrectionGUIEnvironment <- new.env()

as the very first line in this file, and use that for storing these variables.

Martin



?On 5/4/21, 12:36 AM, "Bioc-devel on behalf of Christian Kohler" <bioc-devel-bounces at r-project.org on behalf of christian.kohler at ur.de> wrote:

     Hello BioC community,

     I am the maintainer of the IsoCorrectoRGUI package.
     Starting with April 30th, problems are reported in the Multiple platform
     build/check report for BioC3.13 (Linux, Windows Server as well as macOS)
     though IsoCorrectoRGUIs code base did not change since end of October 2020.

     The relevant part of the error message is shown below:

     * checking examples ... ERROR
     Running examples in ?IsoCorrectoRGUI-Ex.R? failed
     The error most likely occurred in:

      > base::assign(".ptime", proc.time(), pos = "CheckExEnv")
      > ### Name: IsoCorrectionGUI
      > ### Title: Graphical User Interface for IsoCorrectoR
      > ### Aliases: IsoCorrectionGUI
      >
      > ### ** Examples
      >
      >
      >  IsoCorrectionGUI()

     Error in baseEnvRef$continueIsoCorrection <- TRUE :
        cannot add binding of 'continueIsoCorrection' to the base environment
     Calls: IsoCorrectionGUI -> initGUI
     Execution halted

     Frankly speaking, at the moment I have no clue why the CHECK procedure
     fails from one day to the next. Does this reflect a conflict with R
     4.1.0 alpha internals?
     Any insight would be greatly appreciated as the deadline for passing R
     CMD check is in a couple of days.

     If anyone knows why this could be happening or if it is likely to sort
     itself in the next build please let me know.


     Thanks,

     Christian