Re: [HACKERS] TODO list updated

> TODO item comments:
>
> * -SELECT foo UNION SELECT foo is incorrectly simplified to SELECT foo
>
> The above is NOT done.

Fixed.

>
> * prevent primary key that exceeds max index columns [primary]
>
> The above is done as of yesterday.

OK.

>
> * Fix memory leak for expressions[memory](Tom?)
>
> This isn't going to happen for 7.0, looks like :-(

I figured.

>
> * -Allow compression of large fields or a compressed field type
>
> This has to be marked not-done again, unless Jan manages to squeeze
> it back in via the toaster before Feb.

I was optimistic. I will take it off mark.

>
> * Pull requested data directly from indexes, bypassing heap data
>
> I doubt this is ever going to happen --- to make it possible, we'd
> have to store tuple-commit status in index entries as well as in the
> tuples themselves. That would be a substantial space and speed penalty;
> is the potential gain really worth it?

Ingres does this. Not sure if it worth it. Comments?

>
> * -Convert function(constant) into a constant for index use(Tom)
>
> Bernard Frankpitt should get the bulk of the credit for that one, not me.

Updated.

>
> * Allow LIMIT ability on single-table queries that have no ORDER BY to use
> a matching index [limit]
> * Improve LIMIT processing by using index to limit rows processed [limit]
> * Have optimizer take LIMIT into account when considering index scans [limit]
>
> I agree with Hiroshi that these entries are redundant.

Only one remains now.

>
> * -Make index creation use psort code, because it is now faster(Vadim)
>
> I did that, not Vadim.

Vadim had claimed it. You did it. Updated.

>
> * -elog() flushes cache, try invalidating just entries from current xact,
> perhaps using invalidation cache
>
> I don't think this is done?

I thought we fixed this. Hiroshi? I could swear this came in the past
few weeks.

>
> * -Process const = const parts of OR clause in separate pass(Tom)
>
> Again, mostly Frankpitt.

Updated.

>
>
> Some other things I did that aren't mentioned in TODO, but perhaps
> deserve to be shown as 7.0 fixes:
>
> * Interlock to prevent DROP DATABASE on a database with running backends
>
> * Buffer reference counting bugfixes
>
> * Fix libpq bug that causes it to drop backend error message sent
> just before connection closure (ie, any FATAL error message :-().

All added to reliability section.

--
Bruce Momjian | http://www.op.net/~candle
maillist(at)candle(dot)pha(dot)pa(dot)us | (610) 853-3000
+ If your life is a hard drive, | 830 Blythe Avenue
+ Christ can be your backup. | Drexel Hill, Pennsylvania 19026