Обсуждение: Re: [HACKERS] Current open 6.3 issues (fwd)

> > GROUP BY bug of duplicates
> > GROUP BY nulls bug
> > ORDER BY nulls(Vadim?)
> > many OR's exhaust optimizer memory(Vadim?)
> > max tuple length settings(Darren & Peter)
> > pg_user defaults to world-readable until passwords used(Todd)
> > self-join optimizer bug
> > subselects(Vadim)

As far as I know, none of these are ready for 6.3, which is due on
Sunday.  What is the game plan, folks?

And the tools/RELEASE_CHANGES files have not been updated.  I
will do the FAQ, TODO, HISTORY files.

--
Bruce Momjian
maillist@candle.pha.pa.us

Bruce Momjian wrote:
>
> > > GROUP BY bug of duplicates
> > > GROUP BY nulls bug
> > > ORDER BY nulls(Vadim?)
> > > many OR's exhaust optimizer memory(Vadim?)
> > > max tuple length settings(Darren & Peter)
> > > pg_user defaults to world-readable until passwords used(Todd)
> > > self-join optimizer bug
      ^^^^^^^^^^^^^^^^^^^^^^^

Test case, please.

> > > subselects(Vadim)
>
> As far as I know, none of these are ready for 6.3, which is due on
                                                ^^^ beta
> Sunday.  What is the game plan, folks?

I'm working...

Vadim

On Fri, 30 Jan 1998, Bruce Momjian wrote:

> > > GROUP BY bug of duplicates
> > > GROUP BY nulls bug
> > > ORDER BY nulls(Vadim?)
> > > many OR's exhaust optimizer memory(Vadim?)
> > > max tuple length settings(Darren & Peter)
> > > pg_user defaults to world-readable until passwords used(Todd)
> > > self-join optimizer bug
> > > subselects(Vadim)
>
> As far as I know, none of these are ready for 6.3, which is due on
> Sunday.  What is the game plan, folks?

    "bugs" can be fixed up to RELEASE date, which covers 4 of the
above...

> And the tools/RELEASE_CHANGES files have not been updated.  I
> will do the FAQ, TODO, HISTORY files.

    I believe I'm supposed to be handling the RELEASE_CHANGES
file...will take a look into that over the next week or so...its a month
before the RELEASE :)

On Fri, 30 Jan 1998, Vadim B. Mikheev wrote:

> Bruce Momjian wrote:
> >
> > > > GROUP BY bug of duplicates
> > > > GROUP BY nulls bug
> > > > ORDER BY nulls(Vadim?)
> > > > many OR's exhaust optimizer memory(Vadim?)
> > > > max tuple length settings(Darren & Peter)
> > > > pg_user defaults to world-readable until passwords used(Todd)
> > > > self-join optimizer bug
>       ^^^^^^^^^^^^^^^^^^^^^^^
>
> Test case, please.

I'm not sure if this is an example of the self-join bug mentioned above
or just an example of an "exceptional" computing challenge.  I repost
the example so that those of you who understand what's going on can
judge for yourself.

>
> On Fri, 30 Jan 1998, Bruce Momjian wrote:
>
> > > > GROUP BY bug of duplicates
> > > > GROUP BY nulls bug
> > > > ORDER BY nulls(Vadim?)
> > > > many OR's exhaust optimizer memory(Vadim?)
> > > > max tuple length settings(Darren & Peter)
> > > > pg_user defaults to world-readable until passwords used(Todd)
> > > > self-join optimizer bug
> > > > subselects(Vadim)
> >
> > As far as I know, none of these are ready for 6.3, which is due on
> > Sunday.  What is the game plan, folks?
>
>     "bugs" can be fixed up to RELEASE date, which covers 4 of the
> above...
>
> > And the tools/RELEASE_CHANGES files have not been updated.  I
> > will do the FAQ, TODO, HISTORY files.
>
>     I believe I'm supposed to be handling the RELEASE_CHANGES
> file...will take a look into that over the next week or so...its a month
> before the RELEASE :)

This is 6.3 beta.  IMHO, the stuff has to say at least 6.3, if not 6.3
beta by Feb 1.  Don't want beta users saying they have 6.2.1.

--
Bruce Momjian
maillist@candle.pha.pa.us