You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.

Tom Lane 1d33858406 Fix handling of targetlist SRFs when scan/join relation is known empty. ... When we introduced separate ProjectSetPath nodes for application of set-returning functions in v10, we inadvertently broke some cases where we're supposed to recognize that the result of a subquery is known to be empty (contain zero rows). That's because IS_DUMMY_REL was just looking for a childless AppendPath without allowing for a ProjectSetPath being possibly stuck on top. In itself, this didn't do anything much worse than produce slightly worse plans for some corner cases. Then in v11, commit 11cf92f6e rearranged things to allow the scan/join targetlist to be applied directly to partial paths before they get gathered. But it inserted a short-circuit path for dummy relations that was a little too short: it failed to insert a ProjectSetPath node at all for a targetlist containing set-returning functions, resulting in bogus "set-valued function called in context that cannot accept a set" errors, as reported in bug #15669 from Madelaine Thibaut. The best way to fix this mess seems to be to reimplement IS_DUMMY_REL so that it drills down through any ProjectSetPath nodes that might be there (and it seems like we'd better allow for ProjectionPath as well). While we're at it, make it look at rel->pathlist not cheapest_total_path, so that it gives the right answer independently of whether set_cheapest has been done lately. That dependency looks pretty shaky in the context of code like apply_scanjoin_target_to_paths, and even if it's not broken today it'd certainly bite us at some point. (Nastily, unsafe use of the old coding would almost always work; the hazard comes down to possibly looking through a dangling pointer, and only once in a blue moon would you find something there that resulted in the wrong answer.) It now looks like it was a mistake for IS_DUMMY_REL to be a macro: if there are any extensions using it, they'll continue to use the old inadequate logic until they're recompiled, after which they'll fail to load into server versions predating this fix. Hopefully there are few such extensions. Having fixed IS_DUMMY_REL, the special path for dummy rels in apply_scanjoin_target_to_paths is unnecessary as well as being wrong, so we can just drop it. Also change a few places that were testing for partitioned-ness of a planner relation but not using IS_PARTITIONED_REL for the purpose; that seems unsafe as well as inconsistent, plus it required an ugly hack in apply_scanjoin_target_to_paths. In passing, save a few cycles in apply_scanjoin_target_to_paths by skipping processing of pre-existing paths for partitioned rels, and do some cosmetic cleanup and comment adjustment in that function. I renamed IS_DUMMY_PATH to IS_DUMMY_APPEND with the intention of breaking any code that might be using it, since in almost every case that would be wrong; IS_DUMMY_REL is what to be using instead. In HEAD, also make set_dummy_rel_pathlist static (since it's no longer used from outside allpaths.c), and delete is_dummy_plan, since it's no longer used anywhere. Back-patch as appropriate into v11 and v10. Tom Lane and Julien Rouhaud Discussion: https://postgr.es/m/15669-02fb3296cca26203@postgresql.org		7 years ago
..
appendinfo.h	Rename nodes/relation.h to nodes/pathnodes.h.	7 years ago
clauses.h	Build out the planner support function infrastructure.	7 years ago
cost.h	Rename nodes/relation.h to nodes/pathnodes.h.	7 years ago
geqo.h	Rename nodes/relation.h to nodes/pathnodes.h.	7 years ago
geqo_copy.h	Update copyright for 2019	7 years ago
geqo_gene.h	Update copyright for 2019	7 years ago
geqo_misc.h	Update copyright for 2019	7 years ago
geqo_mutation.h	Update copyright for 2019	7 years ago
geqo_pool.h	Update copyright for 2019	7 years ago
geqo_random.h	Update copyright for 2019	7 years ago
geqo_recombination.h	Update copyright for 2019	7 years ago
geqo_selection.h	Update copyright for 2019	7 years ago
inherit.h	Rename nodes/relation.h to nodes/pathnodes.h.	7 years ago
joininfo.h	Rename nodes/relation.h to nodes/pathnodes.h.	7 years ago
optimizer.h	Allow extensions to generate lossy index conditions.	7 years ago
orclauses.h	Rename nodes/relation.h to nodes/pathnodes.h.	7 years ago
paramassign.h	Rename nodes/relation.h to nodes/pathnodes.h.	7 years ago
pathnode.h	Refactor the representation of indexable clauses in IndexPaths.	7 years ago
paths.h	Fix handling of targetlist SRFs when scan/join relation is known empty.	7 years ago
placeholder.h	Rename nodes/relation.h to nodes/pathnodes.h.	7 years ago
plancat.h	Build out the planner support function infrastructure.	7 years ago
planmain.h	Rename nodes/relation.h to nodes/pathnodes.h.	7 years ago
planner.h	Fix handling of targetlist SRFs when scan/join relation is known empty.	7 years ago
prep.h	Rename nodes/relation.h to nodes/pathnodes.h.	7 years ago
restrictinfo.h	Refactor the representation of indexable clauses in IndexPaths.	7 years ago
subselect.h	Rename nodes/relation.h to nodes/pathnodes.h.	7 years ago
tlist.h	Rename nodes/relation.h to nodes/pathnodes.h.	7 years ago