This isn't exactly a bug, but it could be considered unintuitive behavior. Consider this:
CREATE VIEW foo AS SELECT * FROM a; CREATE VIEW foo_v AS SELECT * FROM foo; ALTER TABLE foo ADD COLUMN b INT; The ALTER TABLE statement affects VIEW foo, but the column addition does not propagate to VIEW foo_v. Thus, it makes this deceptive: ... AS SELECT * FROM foo; I ran into this with an application where a real table is accessed if the user is an "admin", while regular users access a view instead. I considered "AS SELECT * FROM foo" to be a promise that all columns from foo would be included in the view, but the promise is broken when ADD COLUMN is applied later on. Would it be a desirable feature to make `CREATE VIEW foo_v AS SELECT * FROM foo;` automatically update the column set when foo's columns change? Instead of the wildcard * being expanded once at CREATE VIEW time, it would (semantically) be expanded every time foo_v is selected on. -- Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org) To make changes to your subscription: http://www.postgresql.org/mailpref/pgsql-hackers
