On Fri, Nov 4, 2016 at 12:15 PM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
Michael Paquier <michael.paquier@gmail.com> writes: > I am passing that down to a committer for review. The patch looks > large, but at 95% it involves diffs in the regression tests, > alternative outputs taking a large role in the bloat.
This is kind of cute, but it doesn't seem to cover very much territory, because it only catches errors that are found in the parse stage. For instance, it fails to cover Franck's original example:
[...]
Maybe it'd be all right to commit this anyway, but I'm afraid the most common reaction would be "why's it give me this info some of the time, but not when I really need it?" I'm inclined to think that an acceptable patch will need to provide context for the plan-time and run-time cases too, and I'm not very sure where would be a sane place to plug in for those cases.