public inbox for [email protected]  
help / color / mirror / Atom feed
pgsql: Fix computation of varnullingrels when translating appendrel Var
4+ messages / 1 participants
[nested] [flat]

* pgsql: Fix computation of varnullingrels when translating appendrel Var
@ 2026-02-20 09:08 Richard Guo <[email protected]>
  0 siblings, 0 replies; 4+ messages in thread

From: Richard Guo @ 2026-02-20 09:08 UTC (permalink / raw)
  To: [email protected]

Fix computation of varnullingrels when translating appendrel Var

When adjust_appendrel_attrs translates a Var referencing a parent
relation into a Var referencing a child relation, it propagates
varnullingrels from the parent Var to the translated Var.  Previously,
the code simply overwrote the translated Var's varnullingrels with
those of the parent.

This was incorrect because the translated Var might already possess
nonempty varnullingrels.  This happens, for example, when a LATERAL
subquery within a UNION ALL references a Var from the nullable side of
an outer join.  In such cases, the translated Var correctly carries
the outer join's relid in its varnullingrels.  Overwriting these bits
with the parent Var's set caused the planner to lose track of the fact
that the Var could be nulled by that outer join.

In the reported case, because the underlying column had a NOT NULL
constraint, the planner incorrectly deduced that the Var could never
be NULL and discarded essential IS NOT NULL filters.  This led to
incorrect query results where NULL rows were returned instead of being
filtered out.

To fix, use bms_add_members to merge the parent Var's varnullingrels
into the translated Var's existing set, preserving both sources of
nullability.

Back-patch to v16.  Although the reported case does not seem to cause
problems in v16, leaving incorrect varnullingrels in the tree seems
like a trap for the unwary.

Bug: #19412
Reported-by: Sergey Shinderuk <[email protected]>
Author: Richard Guo <[email protected]>
Reviewed-by: Tom Lane <[email protected]>
Discussion: https://postgr.es/m/[email protected]
Backpatch-through: 16

Branch
------
master

Details
-------
https://git.postgresql.org/pg/commitdiff/691977d37037f30e39cc65f60e327f678f2cebe9

Modified Files
--------------
src/backend/optimizer/util/appendinfo.c | 12 +++++---
src/test/regress/expected/join.out      | 51 +++++++++++++++++++++++++++++++++
src/test/regress/sql/join.sql           | 24 ++++++++++++++++
3 files changed, 83 insertions(+), 4 deletions(-)



^ permalink  raw  reply  [nested|flat] 4+ messages in thread

* pgsql: Fix computation of varnullingrels when translating appendrel Var
@ 2026-02-20 09:08 Richard Guo <[email protected]>
  0 siblings, 0 replies; 4+ messages in thread

From: Richard Guo @ 2026-02-20 09:08 UTC (permalink / raw)
  To: [email protected]

Fix computation of varnullingrels when translating appendrel Var

When adjust_appendrel_attrs translates a Var referencing a parent
relation into a Var referencing a child relation, it propagates
varnullingrels from the parent Var to the translated Var.  Previously,
the code simply overwrote the translated Var's varnullingrels with
those of the parent.

This was incorrect because the translated Var might already possess
nonempty varnullingrels.  This happens, for example, when a LATERAL
subquery within a UNION ALL references a Var from the nullable side of
an outer join.  In such cases, the translated Var correctly carries
the outer join's relid in its varnullingrels.  Overwriting these bits
with the parent Var's set caused the planner to lose track of the fact
that the Var could be nulled by that outer join.

In the reported case, because the underlying column had a NOT NULL
constraint, the planner incorrectly deduced that the Var could never
be NULL and discarded essential IS NOT NULL filters.  This led to
incorrect query results where NULL rows were returned instead of being
filtered out.

To fix, use bms_add_members to merge the parent Var's varnullingrels
into the translated Var's existing set, preserving both sources of
nullability.

Back-patch to v16.  Although the reported case does not seem to cause
problems in v16, leaving incorrect varnullingrels in the tree seems
like a trap for the unwary.

Bug: #19412
Reported-by: Sergey Shinderuk <[email protected]>
Author: Richard Guo <[email protected]>
Reviewed-by: Tom Lane <[email protected]>
Discussion: https://postgr.es/m/[email protected]
Backpatch-through: 16

Branch
------
REL_17_STABLE

Details
-------
https://git.postgresql.org/pg/commitdiff/bcaf1b5101c44a4549774ef047dafcbf46cfd904

Modified Files
--------------
src/backend/optimizer/util/appendinfo.c | 12 ++++++--
src/test/regress/expected/join.out      | 51 +++++++++++++++++++++++++++++++++
src/test/regress/sql/join.sql           | 24 ++++++++++++++++
3 files changed, 84 insertions(+), 3 deletions(-)



^ permalink  raw  reply  [nested|flat] 4+ messages in thread

* pgsql: Fix computation of varnullingrels when translating appendrel Var
@ 2026-02-20 09:08 Richard Guo <[email protected]>
  0 siblings, 0 replies; 4+ messages in thread

From: Richard Guo @ 2026-02-20 09:08 UTC (permalink / raw)
  To: [email protected]

Fix computation of varnullingrels when translating appendrel Var

When adjust_appendrel_attrs translates a Var referencing a parent
relation into a Var referencing a child relation, it propagates
varnullingrels from the parent Var to the translated Var.  Previously,
the code simply overwrote the translated Var's varnullingrels with
those of the parent.

This was incorrect because the translated Var might already possess
nonempty varnullingrels.  This happens, for example, when a LATERAL
subquery within a UNION ALL references a Var from the nullable side of
an outer join.  In such cases, the translated Var correctly carries
the outer join's relid in its varnullingrels.  Overwriting these bits
with the parent Var's set caused the planner to lose track of the fact
that the Var could be nulled by that outer join.

In the reported case, because the underlying column had a NOT NULL
constraint, the planner incorrectly deduced that the Var could never
be NULL and discarded essential IS NOT NULL filters.  This led to
incorrect query results where NULL rows were returned instead of being
filtered out.

To fix, use bms_add_members to merge the parent Var's varnullingrels
into the translated Var's existing set, preserving both sources of
nullability.

Back-patch to v16.  Although the reported case does not seem to cause
problems in v16, leaving incorrect varnullingrels in the tree seems
like a trap for the unwary.

Bug: #19412
Reported-by: Sergey Shinderuk <[email protected]>
Author: Richard Guo <[email protected]>
Reviewed-by: Tom Lane <[email protected]>
Discussion: https://postgr.es/m/[email protected]
Backpatch-through: 16

Branch
------
REL_16_STABLE

Details
-------
https://git.postgresql.org/pg/commitdiff/ec20a45528bf1731bd532edd51cb68e5f341ac8f

Modified Files
--------------
src/backend/optimizer/util/appendinfo.c | 12 ++++++--
src/test/regress/expected/join.out      | 52 +++++++++++++++++++++++++++++++++
src/test/regress/sql/join.sql           | 24 +++++++++++++++
3 files changed, 85 insertions(+), 3 deletions(-)



^ permalink  raw  reply  [nested|flat] 4+ messages in thread

* pgsql: Fix computation of varnullingrels when translating appendrel Var
@ 2026-02-20 09:08 Richard Guo <[email protected]>
  0 siblings, 0 replies; 4+ messages in thread

From: Richard Guo @ 2026-02-20 09:08 UTC (permalink / raw)
  To: [email protected]

Fix computation of varnullingrels when translating appendrel Var

When adjust_appendrel_attrs translates a Var referencing a parent
relation into a Var referencing a child relation, it propagates
varnullingrels from the parent Var to the translated Var.  Previously,
the code simply overwrote the translated Var's varnullingrels with
those of the parent.

This was incorrect because the translated Var might already possess
nonempty varnullingrels.  This happens, for example, when a LATERAL
subquery within a UNION ALL references a Var from the nullable side of
an outer join.  In such cases, the translated Var correctly carries
the outer join's relid in its varnullingrels.  Overwriting these bits
with the parent Var's set caused the planner to lose track of the fact
that the Var could be nulled by that outer join.

In the reported case, because the underlying column had a NOT NULL
constraint, the planner incorrectly deduced that the Var could never
be NULL and discarded essential IS NOT NULL filters.  This led to
incorrect query results where NULL rows were returned instead of being
filtered out.

To fix, use bms_add_members to merge the parent Var's varnullingrels
into the translated Var's existing set, preserving both sources of
nullability.

Back-patch to v16.  Although the reported case does not seem to cause
problems in v16, leaving incorrect varnullingrels in the tree seems
like a trap for the unwary.

Bug: #19412
Reported-by: Sergey Shinderuk <[email protected]>
Author: Richard Guo <[email protected]>
Reviewed-by: Tom Lane <[email protected]>
Discussion: https://postgr.es/m/[email protected]
Backpatch-through: 16

Branch
------
REL_18_STABLE

Details
-------
https://git.postgresql.org/pg/commitdiff/ed57c207c397a6e6a91e0394cacdc63e0f029173

Modified Files
--------------
src/backend/optimizer/util/appendinfo.c | 12 +++++---
src/test/regress/expected/join.out      | 51 +++++++++++++++++++++++++++++++++
src/test/regress/sql/join.sql           | 24 ++++++++++++++++
3 files changed, 83 insertions(+), 4 deletions(-)



^ permalink  raw  reply  [nested|flat] 4+ messages in thread


end of thread, other threads:[~2026-02-20 09:08 UTC | newest]

Thread overview: 4+ messages (download: mbox mbox.gz follow: Atom feed)
-- links below jump to the message on this page --
2026-02-20 09:08 pgsql: Fix computation of varnullingrels when translating appendrel Var Richard Guo <[email protected]>
2026-02-20 09:08 pgsql: Fix computation of varnullingrels when translating appendrel Var Richard Guo <[email protected]>
2026-02-20 09:08 pgsql: Fix computation of varnullingrels when translating appendrel Var Richard Guo <[email protected]>
2026-02-20 09:08 pgsql: Fix computation of varnullingrels when translating appendrel Var Richard Guo <[email protected]>

This inbox is served by agora; see mirroring instructions
for how to clone and mirror all data and code used for this inbox