On 2/10/2016 6:50 AM, Rob van der Heij wrote: >> > That is, if you specify neither SELECT SECOND nor an EOF item, the >> > run-out cycle is suppressed, even if PRINTONLY EOF is specified. >> > >> > One circumvention is to specify EOF without further specification items. > It's not just EOF - happens with any break level.
Interesting. It's as though the break is never established unless it's tested--almost. If you use the BREAK() function or SELECT SECOND in an IF clause, even if that item's never executed, the break still happens. I notice that the EOF() function also fails to trigger the run-out cycle. Of course, it's useless without SELECT SECOND anyway. I don't see anything about the current behavior of PRINTONLY that anyone could be exploiting. All it does is suppress the output completely. I'm not sure it would be worth fixing just for the EOF case, since without SELECT SECOND or an EOF item the only possible output at EOF is a null record. But if you fix it for a break on an input field, then I'd say fix it for EOF as well (both PRINTONLY and EOF()). ¬R
