Pavel Stehule wrote:
sorry - I dislike this. The design is correct, but it is against to
SQL verbosity. A reader must to thinking about missing tablenames. I
dont't think so it is good solution, because it doesn't solve a
backing compatibility problem - somebody must to fix a function still,
and I think so it is much preferable to fix like:
select (mytbl.mycol + myvar * myparam) as mynewcol from mytbl;
your proposal saves a five chars, but it has a negative impacts on
readability - there should be more tables.
There are no reason to introduce a new concepts - SQL knows a aliases.
Well, going forward, I know I would much rather have to say "mytbl.mycol" than
have to say "myfunc.myparam". And I certainly would want to expect that when
one says "... as foo" that this "foo" is treated as a declaration unambiguously
and is never substituted for some parameter or there be other grief as I seem to
recall having in 8.4. -- Darren Duncan
--
Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-hackers
