On Mon, 13 Jun 2005, Chris Bergstresser wrote:
> Hi all --
>
> I'm seeing the following commands reliably produce a crash in RGui,
> version 2.0.1, for both my home and office machine:
>
> > rm(list = ls(all = TRUE));
> > load("dataset.R");
> > library("wle");
> > data.wle = wle.lm(abortion ~ year * lib.con + age + gender +
> + urbanism + census + income + church.att + children + educ +
> + religion.imp, data = data.set);
>
> dataset.R is moderately sized (about 200k compressed), and the
> command works just fine with "lm" rather than "wle.lm".
Since I'm not
> sure where the bug lies -- in wle, in R, or in RGui -- I'm not sure
> where I should report this, or if it's already been reported.
> What should I do?
There is good advice in the posting guide. Following that
1) Re-do the tests in the current version of R, preferably a beta of
2.1.1.
2) Read the rw-FAQ, do the debugging reported there (with Dr MinGW or gdb)
and find where it is crashing. (This is very likely to be in wle.)
If it is in wle, send a report to the maintainer. If it is in R, send a
report to R-bugs. In either case, supply enough data to reproduce the
problem.
--
Brian D. Ripley, ripley at stats.ox.ac.uk
Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/
University of Oxford, Tel: +44 1865 272861 (self)
1 South Parks Road, +44 1865 272866 (PA)
Oxford OX1 3TG, UK Fax: +44 1865 272595