Hi Maciej Thanks for the thorough explanation of the problem.
> My proposal therefore is to simplify the return path from > remote_expect and except from keeping the current special handling > of the error exception intact, pass all the other conditions expect > might have produced up to the caller. Below is the proposed > implementation. It makes a loop like above behave as expected. I agree that remote_expect should behave as similar to expect as possible to adhere to the principle of least surprise. I will apply your patch now, thanks. Cheers, Ben _______________________________________________ Bug-dejagnu mailing list [email protected] http://lists.gnu.org/mailman/listinfo/bug-dejagnu
