sql99 compat list

On Mon, Aug 11, 2003 at 09:40:25AM +0800, Christopher Kings-Lynne wrote:

Has anyone reviewed the compatibility list for 7.4 yet?

I seem to remember something about us having the unique predicate now or
something?

I hacked up a really simplistic implementation of it, but it wasn't
included in 7.4 since it was only half-baked.

cursors?

I believe I updated the list for the new cursor features, but let me know
if I missed anything.

-Neil

Christopher Kings-Lynne writes:

Has anyone reviewed the compatibility list for 7.4 yet?

Yes.

I seem to remember something about us having the unique predicate now or
something?

It's not there yet.

Array support is now better,

But not good enough.

and cursors?

Same.

String ops for LOBs?

We'd need to have CLOB and BLOB data types in the first place.

--
Peter Eisentraut peter_e@gmx.net

Christopher Kings-Lynne wrote:

Has anyone reviewed the compatibility list for 7.4 yet?

http://developer.postgresql.org/docs/postgres/unsupported-features-sql99.html

I seem to remember something about us having the unique predicate now or
something? Array support is now better, and cursors? String ops for LOBs?

Chris

---------------------------(end of broadcast)---------------------------
TIP 7: don't forget to increase your free space map settings

Doesn't appear to be on this list, but shouldn't the fact that PSQL
translates undoublequoted identifiers to lower instead of upper case be
mentioned? Also, I think a pretty simple workaround would be to have
PSQL search for the upper case identifier, and if not found, search for
the lower case. This should allow a migration path while the tools and
functions are being translated (and should, perhaps, stay around forever
as a session option for old databases)

Shachar

--
Shachar Shemesh
Open Source integration consultant
Home page & resume - http://www.shemesh.biz/

Shachar Shemesh <psql@shemesh.biz> writes:

Also, I think a pretty simple workaround would be to have
PSQL search for the upper case identifier, and if not found, search for
the lower case. This should allow a migration path while the tools and
functions are being translated (and should, perhaps, stay around forever
as a session option for old databases)

You're assuming that we plan to change this. We don't. Most of us
prefer to look at lower-case identifiers.

regards, tom lane

Tom Lane wrote:

Shachar Shemesh <psql@shemesh.biz> writes:

Also, I think a pretty simple workaround would be to have
PSQL search for the upper case identifier, and if not found, search for
the lower case. This should allow a migration path while the tools and
functions are being translated (and should, perhaps, stay around forever
as a session option for old databases)

You're assuming that we plan to change this. We don't. Most of us
prefer to look at lower-case identifiers.

regards, tom lane

While I cannot argue with personal preferences, of course, I will try to
point out two things:
1. It's an SQL incompatibility, and therefor must be documented (at the
very least).
2. It is causing pains when performing migrations from other databases

At the very least, a session option that will allow me to search for
upper case if a lowercase identifier was not found would be apretiated,
optionally with a warning (so I can track down the places in the
migrated application that are still misbehaving, and fix them).

Shachar

--
Shachar Shemesh
Open Source integration consultant
Home page & resume - http://www.shemesh.biz/