Yasushi SHOJI <yasushi.shoji@xxxxxxxxx> writes: > The patch (actually, I've tested the one in pu, 2e9fdc795cb27) avoids > the seg fault for sure, but the question is: > > When does the list allowed to contain NULLs? A very legitimate question. With the proposed log message alone, it is even tempting to declare that the change may merely be sweeping the issue under the rug. A bit better explanation is needed, at least.