any size now.

dynamically with PQExpBuffer.

mysteriously disappearing.

ie.  \d+ will only ever show the comment for the current database --
which is appropriate since it can only pull comments from the current
database.

Won't break pgadmin functionality as it enforces this behaviour already.


I didn't find any regression tests for COMMENT.

Rod Taylor

no objections.

Joe Conway

left a stub for a future "ALTER RULE RENAME" but did not write that one
yet. Bruce, if you want to add my name for for that I'll take it and do
it later.

Joe Conway

default values using ALTER TABLE, on both views and tables.

(You'll need to apply the default-values-for-views patch that I sent
to -patches earlier for the regression tests to pass.)

Neil Conway <neilconway@rogers.com>

postgres command line utilites e.g. supports -U, -p, -h, -?, -v, password
prompt and has a "test mode". In test mode, no large objects are removed,
just reported.

Mario Weilguni

unused function arguments, and makes the TableInfo struct slightly
smaller by removing an unnecessary member.

Neil Conway <neilconway@rogers.com>

Apparently, you need to make two calls to appendPQExpBuffer() to
use fmtId() twice, because it uses a static buffer (thanks for
spotting this Tom).

Another revision of the patch is attached.

Neil Conway <neilconway@rogers.com>

The attached patch allows views to have default values. You can't
specify a default value within a CREATE VIEW statement, it must be
done using ALTER TABLE ... ALTER COLUMN ... SET DEFAULT after the
view has already been created. Most of the hard work was done by
Tom Lane, I just patched pg_dump and updated the documentation.

Neil Conway <neilconway@rogers.com>

Appears I forgot to update the docs earlier.

Rod Taylor

Please note I have no way to test this.  Docbook doesn't like my
cygwin setup.

--
Rod Taylor

Zhenbang Wei

objections.

Major changes:
   - removed cursor wrap around input sql to allow for remote
     execution of INSERT/UPDATE/DELETE
   - dblink now returns a resource id instead of a real pointer
   - added several utility functions

I'm still hoping to add explicit cursor open/fetch/close support before
7.3 is released, but I need a bit more time on that.

On a somewhat unrelated topic, I never got any feedback on the
unknownin/out patch and the mb_substring patch. Is there anything else I
need to do to get those applied?

Joe Conway

work on all win9x machines, so i made it go thru a l ookup table
instead, using the DLL as last resort.  I also moved this out of the
fe-misc.c file because of the size of the lookup ta ble. Who knows, we
might add more other win32 specific code there in the future.

I also fixed a small typo in the pg_config.h.win32 that made the
compiler compla in about the gnu snprintf declaration.

I tried to make this patch with psql coding style. I've successfully
tested this on win2k and win98 and it works fine (i.e. the mes sage
shows on win98 too, it didn't with the old implementation).

Magnus Naeslund

BAD:  INSERT INTO tab (col1, col2) VALUES ('val1');
GOOD: INSERT INTO tab (col1, col2) VALUES ('val1', 'val2');

Regress tests against DEFAULT and normal values as they're managed
slightly different.

Rod Taylor

Mario Weilguni

looking for places that assume UNKNOWN == TEXT. One of those was the
"SET" type in pg_type.h, which was using textin/textout. This one I took
care of in this patch. The other suspicious place was in
string_to_dataum (which is defined in both selfuncs.c and indxpath.c). I
wasn't too sure about those, so I left them be.

Joe Conway

Slackware 8), and perhaps on other Pythons, haven't checked.  Something in
the _pg.connect() call isn't working.  I think the problem stems from the
fact that 'host' is a named parameter of both _pg.connect and pgdb.connect,
and so Python treats it as a variable assignment, not a named parameter.

Uses non-named parameters.

Andrew Johnson

the new verion compiled on SCO Openserver 5.0.5 and Unixware 7.1.1

Nicolas Bazin

Ulrich Neumann

[ Patch comments in three pieces.]

Attached is a pacth against 7.2 which adds locale awareness to the
character classes of the regular expression engine.

...

> > I still think the xdigit class could be handled the same way the digit
> > class is (by enumeration rather than using the isxdigit function). That
> > saves you a cicle, and I don't think there's any loss.
>
> In fact, I will email you when I apply the original patch.

I miss that case :-(. Here is the pached patch.

...

Here is a patch which addresses Tatsuo's concerns (it does return an
static struct instead of constructing it).

src/include/pg_config.h.win32.  Certainly if we have INDEX_MAX_KEYS in
there, we should have FUNC_MAX_ARGS too.

Report from Chris Ryan

> * Improve ability to display optimizer analysis using OPTIMIZER_DEBUG

> * -Use faster flex flags for performance improvement (Peter E)

> * -Allow user to control trigger firing order (Tom)
> * -Compile in syslog functionaility by default (Tatsuo)

> * -Allow user to control trigger firing order
> * Add ALTER TRIGGER ... RENAME

UINT64CONST, since unsigned was what it wanted anyway.  Centralize macro
definitions into c.h.

2) Handle timestamp without time zone.
3) Improve SQLForeignKeys() in multibyte mode.