Really fix the ambiguity in REFRESH MATERIALIZED VIEW CONCURRENTLY.
authorTom Lane
Sat, 7 Aug 2021 17:29:32 +0000 (13:29 -0400)
committerTom Lane
Sat, 7 Aug 2021 17:29:32 +0000 (13:29 -0400)
commitba9f665a4413f855bbf4b544481db326f7b9bd73
tree9601e92f5d35df8ef87a3c0c8273716aee11b4fd
parentda188b993450c824b03e7dca18c27e9f4c04754f
Really fix the ambiguity in REFRESH MATERIALIZED VIEW CONCURRENTLY.

Rather than trying to pick table aliases that won't conflict with
any possible user-defined matview column name, adjust the queries'
syntax so that the aliases are only used in places where they can't be
mistaken for column names.  Mostly this consists of writing "alias.*"
not just "alias", which adds clarity for humans as well as machines.
We do have the issue that "SELECT alias.*" acts differently from
"SELECT alias", but we can use the same hack ruleutils.c uses for
whole-row variables in SELECT lists: write "alias.*::compositetype".

We might as well revert to the original aliases after doing this;
they're a bit easier to read.

Like 75d66d10e, back-patch to all supported branches.

Discussion: https://postgr.es/m/2488325.1628261320@sss.pgh.pa.us
src/backend/commands/matview.c
src/test/regress/expected/matview.out
src/test/regress/sql/matview.sql