Dropping a schema

Oliver Elphick <olly@lfix.co.uk> writes:

olly=# drop schema testing;
NOTICE: table testing.testa depends on schema testing
ERROR: Cannot drop schema testing because other objects depend on it
Use DROP ... CASCADE to drop the dependent objects too

This seems a little over-restrictive to me.

It's per spec: SQL92 saith

3) If RESTRICT is specified, then S shall not contain any per-
sistent base tables, global temporary tables, created local
temporary tables, views, domains, assertions, character sets,
collations, or translations.

Note: If CASCADE is specified, then such objects will be dropped
by the effective execution of the SQL schema manipulation state-
ments specified in the General Rules of this Subclause.

Also, it seems the safest behavior to me. "rmdir dir" won't remove a
nonempty directory; isn't that a pretty close analogy?

regards, tom lane

On Wed, 2002-08-21 at 15:02, Tom Lane wrote:

Oliver Elphick <olly@lfix.co.uk> writes:

olly=# drop schema testing;
NOTICE: table testing.testa depends on schema testing
ERROR: Cannot drop schema testing because other objects depend on it
Use DROP ... CASCADE to drop the dependent objects too

This seems a little over-restrictive to me.

It's per spec: SQL92 saith

...

Also, it seems the safest behavior to me. "rmdir dir" won't remove a
nonempty directory; isn't that a pretty close analogy?

Not really, seeing that you can't say "mkdir directory (containing these
files)". An implicit cascade *inside* the schema seems an appropriate
parallel to "CREATE SCHEMA ... (CREATE TABLE ...)". After all, we don't
have to say "DROP TABLE ... CASCADE" because the table has rows in it!

But if that's what the spec says...

--
Oliver Elphick Oliver.Elphick@lfix.co.uk
Isle of Wight, UK
http://www.lfix.co.uk/oliver
GPG: 1024D/3E1D0C1C: CA12 09E0 E8D5 8870 5839 932A 614D 4C34 3E1D 0C1C
========================================
"For the Lord himself shall descend from heaven with a
shout, with the voice of the archangel, and with the
trump of God; and the dead in Christ shall rise first;
Then we which are alive and remain shall be caught
up together with them in the clouds, to meet the Lord
in the air; and so shall we ever be with the Lord."
I Thessalonians 4:16,17

Oliver Elphick <olly@lfix.co.uk> writes:

On Wed, 2002-08-21 at 15:02, Tom Lane wrote:

Also, it seems the safest behavior to me. "rmdir dir" won't remove a
nonempty directory; isn't that a pretty close analogy?

Not really, seeing that you can't say "mkdir directory (containing these
files)". An implicit cascade *inside* the schema seems an appropriate
parallel to "CREATE SCHEMA ... (CREATE TABLE ...)". After all, we don't
have to say "DROP TABLE ... CASCADE" because the table has rows in it!

Hm. I could see an argument for being willing to auto-drop stuff that
had been made that way (inside CREATE SCHEMA) but not stuff that was
made by separate commands. But the spec doesn't seem to make any such
distinction: RESTRICT is RESTRICT. In any case, I like the behavior as
it is, so I'm not gonna go out of my way to change it...

regards, tom lane