Re: Optimizing nbtree ScalarArrayOp execution, allowing multi-column ordered scans, skip scan

Peter Geoghegan <pg(at)bowt(dot)ie> writes:
> On Sun, Apr 7, 2024 at 8:48 PM Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
>> Coverity pointed out something that looks like a potentially live
>> problem in 5bf748b86:
>> ... which certainly makes it look like array->scan_key could be
>> a null-pointer dereference.

> But the "Assert(xform[j].ikey == array->scan_key)" assertion is
> located in a block where it's been established that the scan key (the
> one stored in xform[j] at this point in execution) must have an array.

> This is probably very hard for tools like Coverity to understand.

It's not obvious to human readers either ...

> Would Coverity stop complaining if I just removed the assertion? I
> could just do that, I suppose, but that seems backwards to me.

Perhaps this'd help:

- Assert(xform[j].ikey == array->scan_key);
+ Assert(array && xform[j].ikey == array->scan_key);

If that doesn't silence it, I'd be prepared to just dismiss the
warning.

Some work in the comment to explain why we must have an array here
wouldn't be out of place either, perhaps.

regards, tom lane