On Fri, May 7, 2021 at 7:08 AM Tom Lane <t...@sss.pgh.pa.us> wrote: > > Alvaro Herrera <alvhe...@alvh.no-ip.org> writes: > > On 2021-May-06, Peter Smith wrote: > >> PSA v3 of the patch. Same as before, but now also renames the global > >> variable from "wrconn" to "lrep_worker_wrconn". > > > I think there are two patches here -- the changes to > > AlterSubscription_refresh are a backpatchable bugfix, and the rest of it > > can just be applied to master. > > The rename of that variable is just cosmetic, true, but I'd still be > inclined to back-patch it. If we don't do so then I'm afraid that > future back-patched fixes might be bitten by the same confusion, > possibly introducing new real bugs. > > Having said that, keeping the two aspects in separate patches might > ease review and testing.
Done. > > > In my mind we make a bit of a distinction for global variables by using > > CamelCase rather than undercore_separated_words. > > I think it's about 50/50, TBH. I'd stick with whichever style is > being used in nearby code. > The nearby code was a random mixture of Camels and Snakes, so instead of flipping a coin I went with the suggestion from Alvaro. ~~ PSA v4 of the patch. 0001 - Fixes the AlterSubscription_refresh as before. 0002 - Renames the global var "wrconn" -> "LogRepWorkerWalRcvConn" as suggested. ------ Kind Regards, Peter Smith Fujitsu Australia
v4-0001-AlterSubscription_refresh-Use-stack-variable-for-.patch
Description: Binary data
v4-0002-Rename-the-logical-replication-global-wrconn.patch
Description: Binary data
