On Wed, Mar 14, 2018 at 10:50:38AM +0900, Michael Paquier wrote: > For the rest, which basically concerns psql, I have been thinking that > actually using 2) would be the most painful approach, still something > which does not impact the user experience, while 3) is easier to > back-patch by minimizing the code footprint and avoids also any kind of > future problems.
Actually, there is also the case of pgbench, where there are two items to be careful about: 1) Tweak correctly parameters in built-in benchmark queries. 2) Make pgbench gain an extra option allowing one to run the benchmark on a wanted schema. 1) is essential to do, 2) perhaps less as custom benchmarks can always be designed so the query strings are secured. Most users don't run pgbench on multi-user, untrusted systems as well (right?), giving less value to 2). -- Michael
signature.asc
Description: PGP signature
