]> git.ipfire.org Git - thirdparty/postgresql.git/commit
Fix parse_cte.c's failure to examine sub-WITHs in DML statements.
authorTom Lane <tgl@sss.pgh.pa.us>
Sat, 5 Apr 2025 19:01:33 +0000 (15:01 -0400)
committerTom Lane <tgl@sss.pgh.pa.us>
Sat, 5 Apr 2025 19:01:33 +0000 (15:01 -0400)
commit754a3d82d79cc3d7040cfadaf9c8cd1579684ee6
tree36f1d07eb23e25f9ff6914e73f919fe027d6a790
parenta749c6f18fbacd05f432cd29f9e7294033bc666f
Fix parse_cte.c's failure to examine sub-WITHs in DML statements.

makeDependencyGraphWalker thought that only SelectStmt nodes could
contain a WithClause.  Which was true in our original implementation
of WITH, but astonishingly we missed updating this code when we added
the ability to attach WITH to INSERT/UPDATE/DELETE (and later MERGE).
Moreover, since it was coded to deliberately block recursion to a
WithClause, even updating raw_expression_tree_walker didn't save it.

The upshot of this was that we didn't see references to outer CTE
names appearing within an inner WITH, and would neither complain about
disallowed recursion nor account for such references when sorting CTEs
into a usable order.  The lack of complaints about this is perhaps not
so surprising, because typical usage of WITH wouldn't hit either case.
Still, it's pretty broken; failing to detect recursion here leads to
assert failures or worse later on.

Fix by factoring out the processing of sub-WITHs into a new function
WalkInnerWith, and invoking that for all the statement types that
can have WITH.

Bug: #18878
Reported-by: Yu Liang <luy70@psu.edu>
Author: Tom Lane <tgl@sss.pgh.pa.us>
Discussion: https://postgr.es/m/18878-a26fa5ab6be2f2cf@postgresql.org
Backpatch-through: 13
src/backend/parser/parse_cte.c
src/test/regress/expected/with.out
src/test/regress/sql/with.sql