Apply fixes for problems with dropped columns whose types have also been
authorTom Lane <tgl@sss.pgh.pa.us>
Mon, 12 May 2003 00:17:03 +0000 (00:17 +0000)
committerTom Lane <tgl@sss.pgh.pa.us>
Mon, 12 May 2003 00:17:03 +0000 (00:17 +0000)
commitd97c9b366273a49f0469184df9dfb3a312b2f3ff
tree38e38e6c1d443aa69fb7418a2b0685e10f52de36
parenta7c50cd92b17885d926fdf4f0df084849f285d58
Apply fixes for problems with dropped columns whose types have also been
dropped.  The simplest fix for INSERT/UPDATE cases turns out to be for
preptlist.c to insert NULLs of a known-good type (I used INT4) rather
than making them match the deleted column's type.  Since the representation
of NULL is actually datatype-independent, this should work fine.
I also re-reverted the patch to disable the use_physical_tlist optimization
in the presence of dropped columns.  It still doesn't look worth the
trouble to be smarter, if there are no other bugs to fix.
Added a regression test to catch future problems in this area.
src/backend/catalog/heap.c
src/backend/optimizer/plan/createplan.c
src/backend/optimizer/prep/preptlist.c
src/backend/optimizer/util/plancat.c
src/test/regress/expected/alter_table.out
src/test/regress/sql/alter_table.sql