Request a Demo
Home   >   mailing lists

Re: [oauth] Stabilize the libpq-oauth ABI (and allow alternative implementations?) - Mailing list pgsql-hackers

From Jacob Champion
Subject Re: [oauth] Stabilize the libpq-oauth ABI (and allow alternative implementations?)
Date
Msg-id CAOYmi+nRmt+K+ZVNztKFv=LuasDCCSsXLihZB6BEkVeroW8EvA@mail.gmail.com
Whole thread
In response to Re: [oauth] Stabilize the libpq-oauth ABI (and allow alternative implementations?)  (Jacob Champion <jacob.champion@enterprisedb.com>)
Responses Re: [oauth] Stabilize the libpq-oauth ABI (and allow alternative implementations?)
Re: [oauth] Stabilize the libpq-oauth ABI (and allow alternative implementations?)
List pgsql-hackers
Tree view 
On Tue, Mar 3, 2026 at 2:08 PM Jacob Champion
<jacob.champion@enterprisedb.com> wrote:
> I don't plan to block the other patches on -0006, since -0004 is
> blocking PGOAUTHCAFILE.

v5-0001 and -0002 are now committed. Attached is a v6 rebase with some
minor self-review fixups and the removal of ASan from the poisoning
experiment.

v6-0001 through -0003 are targeted for commit next. I'll leave -0004
for later; I plan to ask for some committer review, since it's pretty
unusual code for libpq, but it needs a real commit message and some
better docs first.

As for -0005: I think I'll cherry-pick only the ability for the
libpq_oauth_init entry to be optional, marking any flows that don't
provide it as user-defined, to make it possible for v19 developers to
help us iterate on v20 if they want.

Thanks,
--Jacob
Attachment

pgsql-hackers by date:

Previous
From: Chao Li
Date:
Subject: Re: [BUG + PATCH] DSA pagemap out-of-bounds in make_new_segment odd-sized path
Next
From: Jacob Champion
Date:
Subject: Re: [PATCH] Add PQgetThreadLock() to expose the Kerberos/Curl mutex