On 18.02.21 17:11, David G. Johnston wrote:
The OP was doing a course based on Oracle and was confused regarding our behavior. The documentation failed to help me provide a useful response, so I'd agree there is something here that needs reworking if not outright fixing.
According to the piece of the standard that I posted, the sensitivity behavior here is implementation-dependent (not even -defined), so both implementations are correct.
But the poster was apparently also confused by the same piece of documentation.
If you consider the implementation of MVCC in PostgreSQL, then the current behavior makes sense. I suspect that this consideration was much more interesting for older system with locking-based concurrency and where "read uncommitted" was a real thing. With the current system, insensitive cursors are essentially free and sensitive cursors would require quite a bit of effort to implement.
