Hi hackers!
In this thread, I want to promote entries from CommitFest that require review.
I have scanned through the bugs, clients, and documentation sections, and here
is my take on the current situation. All of these threads are currently in the
"Needs review" state and were marked by the patch author as targeting version
17.
Bugs:
* LockAcquireExtended improvement
Not really a bug, but rather a limitation. Thread might be of interest for
a reviewer who wants to dive into heavy locks. Some input from Robert. I doubt
this improvement will land into 17.
* Dump-restore loosing 'attnotnull' bit for DEFERRABLE PRIMARY KEY column(s)
3-liner fix, with input from Alvaro.
* Avoid deadlock and concurrency during orphan temp table removal
A real deadlock from production. 0 prior external review, hurry up to be
first reviewer :) The patch is small, yet need a deep understanding of lock
protocols of different combination of objects.
* Explicitly show dependent types as extension members
Some issues on the tip of FDW and types subsystems. Fix by Tom, not
backpatchable. If you want better understanding of extendebility and type
dependencies - take this for review. IMO most probably will be in 17, just need
some extra eyes.
* initdb's -c option behaves wrong way
Fundamental debate, might seems much like tabs vs spaces (strncmp vs
strncasecmp). Patch by Tom, perfect as usual, needs agreement from someone
who's already involved.
Clients:
* vacuumdb/clusterdb/reindexdb: allow specifying objects to process in all
databases
Nice feature, some review from Kyotaro Horiguchi, but need more.
* Support for named parsed statement in psql
Some review by Jelte was done, but seem to require more attention.
* Extend pgbench partitioning to pgbench_history
Tomas Vondra explressed some backward comparability concerns, Melanie
questioned implementation details. I doubt this will land into 17, but
eventually might be part of a famous "sort of TPC-B".
* psql: Allow editing query results with \gedit
There's a possible new nice psql feature, but as of January architectural
discussion was still going on. It does not seem like it will be in 17, unless
some agreement emerge.
Documentation:
* SET ROLE documentation improvement
Thin matters of superuser documentation. Nathan signed up as a committer,
but the thread has no updates for some months.
* Add detail regarding resource consumption wrt max_connections
Cary Huang reviewd the patch, but the result is inconclusive.
* Quick Start Guide to PL/pgSQL and PL/Python Documentation
Some comments from Pavel and Li seem like were not addressed.
* Simplify documentation related to Windows builds
Andres had a couple of notes that were addressed by the author.
* PG DOCS - protocol specifying boolean parameters without values.
Small leftover in a relatevely old documentation thread. Needs a bump from
someone in the thread.
* Documentation: warn about two_phase when altering a subscription
Amit LGTMed, I've added him to cc of the followup.
Stay tuned for other CF sections.
Best regards, Andrey Borodin, learning how to be a CFM.
