I didn't know that a seg fault was an operating system error message (in fact, I don't really know what a seg fault is... I'm not a C programmer, just plain Mr Joe Q Public, R user).
I would say the FAQ makes sense only if you know what kind of error messages operating systems are likely to report. Mentioning seg faults explicitly in FAQ-9.1 might be helpful for nonprogrammers like me. Something like "a seg fault is always a bug" would be good.
best wishes
robin
On Tue, 14 Oct 2003, Robin Hankin wrote:
Hello again
thanks for this. What I should have asked is, should one always report repeatable seg faults (even if functions are called with inappropriate arguments)?
Yes. Reproducible segfaults that occur `in the wild' are fairly important bugs even if they result from errors.
Segfaults from trying to break the system are a much lower priority, but even they should probably be reported.
The FAQ (9.1) says If R executes an illegal instruction, or dies with an operating system error message that indicates a problem in the program (as opposed to something like "disk full"), then it is certainly a bug. and goes on to explain that the only time this doesn't apply is when you call compiled code with the wrong arguments or call your own compiled code.
-thomas
______________________________________________ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
______________________________________________ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
