On Thu, Sep 3, 2009 at 10:16 AM, Gabor Grothendieck<ggrothendi...@gmail.com> wrote: > The danger is that it could introduce bugs into the process > of reporting bugs.
If you mean that sessionInfo() will result in an error, that is easily solved by a tryCatch() statement. Timeouts are harder. One can also consider providing an sessionInfo(..., extras=TRUE) option. The purpose is to save the maintainer (and those who try to help troubleshooting) time. There are tons of time spend on sending follow up requests on basic things. We took a great leap when introducing sessionInfo() and think we can take it further. /Henrik > > On Thu, Sep 3, 2009 at 1:05 PM, Henrik Bengtsson<h...@stat.berkeley.edu> > wrote: >> Hi, >> >> sessionInfo() has been proven really useful, but you still often have >> to ask for additional information in order to help troubleshooting. >> For instance, for troubleshooting the aroma.affymetrix, it is very >> helpful to know what the current working directory is, for other >> packages certain system environment variables might be useful and so >> on. >> >> One solution could be to provide support for hook functions of >> sessionInfo(), which package then can register and append to the >> 'sessionInfo' list structure. Does this make sense? >> >> /Henrik >> >> ______________________________________________ >> R-devel@r-project.org mailing list >> https://stat.ethz.ch/mailman/listinfo/r-devel >> > ______________________________________________ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
