On Fri, Mar 25, 2022 at 7:23 PM Tom Lane <t...@sss.pgh.pa.us> wrote: > > Ashutosh Bapat <ashutosh.bapat....@gmail.com> writes: > > All these functions are too low level to be helpful to know. Knowing > > the caller might actually give a hint as to where the unknown node > > originated from. We may get that from the stack trace if that's > > available. But if we could annotate the error with error_context that > > will be super helpful. > > Is it really that interesting? If function X lacks coverage for > node type Y, then X is what needs to be fixed. The exact call > chain for any particular failure seems of only marginal interest, > certainly not enough to be building vast new infrastructure for. >
I don't think we have tests covering all possible combinations of expression trees. Code coverage reports may not reveal this. I have encountered flaky "unknown expression type" errors. Always ended up changing code to get the stack trace. Having error context helps. -- Best Wishes, Ashutosh Bapat
