Christian Daudt <csd_ob@daudt.org> writes:
> I agree completely - that's why I said my patch was only addressing the
> consequence, not the cause . But the fact is that, regardless of the problem
> that caused the code to call stringbuf_set with null (which remains to be
> fixed) , svn_stringbuf_set shouldn't coredump because it received a value it
> doesn't like. It should either set the string to empty (which is what my
> patch did) or return an error and then get the callers to test for that
> error.
Different projects have different conventions for this. In
Subversion, we seem to have decided that it's okay for code to seg
fault in "can't happen" situations, because if the can't-happen ever
*does* happen, it's a bug we must fix anyway :-). A verbose error and
exit would only be marginally friendlier to the user -- it wouldn't
really get them any closer to a completed operation than the seg fault
does.
-K
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Thu Jun 19 23:29:35 2003