pg.ddx.io pgsql-committers@postgresql.org mailing list archivehelp / color / mirror / Atom feed
pgsql: Teach pg_dump about the new pg_subscription.subrunasowner option 2+ messages / 1 participants [nested] [flat]
* pgsql: Teach pg_dump about the new pg_subscription.subrunasowner option @ 2023-10-29 16:56 Tom Lane <tgl@sss.pgh.pa.us> 0 siblings, 0 replies; 2+ messages in thread From: Tom Lane @ 2023-10-29 16:56 UTC (permalink / raw) To: pgsql-committers@lists.postgresql.org Teach pg_dump about the new pg_subscription.subrunasowner option. Among numerous other oversights, commit 482675987 neglected to fix pg_dump to dump this new subscription option. Since the new default is "false" while the previous behavior corresponds to "true", this would cause legacy subscriptions to silently change behavior during dump/reload or pg_upgrade. That seems like a bad idea. Even if it was intended, failing to preserve the option once set in a new installation is certainly not OK. While here, reorder associated stanzas in pg_dump to match the field order in pg_subscription, in hopes of reducing the impression that all this code was written with the aid of a dartboard. Back-patch to v16 where this new field was added. Philip Warner (cosmetic tweaks by me) Discussion: https://postgr.es/m/20231027042539.01A3A220F0A@thebes.rime.com.au Branch ------ master Details ------- https://git.postgresql.org/pg/commitdiff/5ba4cc309095eb1a947961e66b9c5156efc8ffca Modified Files -------------- src/bin/pg_dump/pg_dump.c | 64 ++++++++++++++++++++++++++++------------------- src/bin/pg_dump/pg_dump.h | 9 ++++--- 2 files changed, 43 insertions(+), 30 deletions(-) ^ permalink raw reply [nested|flat] 2+ messages in thread
* pgsql: Teach pg_dump about the new pg_subscription.subrunasowner option @ 2023-10-29 16:56 Tom Lane <tgl@sss.pgh.pa.us> 0 siblings, 0 replies; 2+ messages in thread From: Tom Lane @ 2023-10-29 16:56 UTC (permalink / raw) To: pgsql-committers@lists.postgresql.org Teach pg_dump about the new pg_subscription.subrunasowner option. Among numerous other oversights, commit 482675987 neglected to fix pg_dump to dump this new subscription option. Since the new default is "false" while the previous behavior corresponds to "true", this would cause legacy subscriptions to silently change behavior during dump/reload or pg_upgrade. That seems like a bad idea. Even if it was intended, failing to preserve the option once set in a new installation is certainly not OK. While here, reorder associated stanzas in pg_dump to match the field order in pg_subscription, in hopes of reducing the impression that all this code was written with the aid of a dartboard. Back-patch to v16 where this new field was added. Philip Warner (cosmetic tweaks by me) Discussion: https://postgr.es/m/20231027042539.01A3A220F0A@thebes.rime.com.au Branch ------ REL_16_STABLE Details ------- https://git.postgresql.org/pg/commitdiff/67738dbf9c9232c8dc1892f970a0bf89617058a6 Modified Files -------------- src/bin/pg_dump/pg_dump.c | 64 ++++++++++++++++++++++++++++------------------- src/bin/pg_dump/pg_dump.h | 9 ++++--- 2 files changed, 43 insertions(+), 30 deletions(-) ^ permalink raw reply [nested|flat] 2+ messages in thread
end of thread, other threads:[~2023-10-29 16:56 UTC | newest] Thread overview: 2+ messages (download: mbox mbox.gz follow: Atom feed) -- links below jump to the message on this page -- 2023-10-29 16:56 pgsql: Teach pg_dump about the new pg_subscription.subrunasowner option Tom Lane <tgl@sss.pgh.pa.us> 2023-10-29 16:56 pgsql: Teach pg_dump about the new pg_subscription.subrunasowner option Tom Lane <tgl@sss.pgh.pa.us>
This inbox is served by DDX for PostgreSQL; see mirroring instructions for how to clone and mirror all data and code used for this inbox