Re: WIP: Covering + unique indexes.
Teodor Sigaev <teodor@sigaev.ru>
Thank you, pushed
Peter Geoghegan wrote:
> On Wed, Apr 18, 2018 at 10:47 PM, Teodor Sigaev <teodor@sigaev.ru> wrote:
>> Thank you, pushed.
>
> Thanks.
>
> I saw another preexisting issue, this time one that has been around
> since 2007. Commit bc292937 forgot to remove a comment above
> _bt_insertonpg() (the 'afteritem' stuff ended up being moved to the
> bottom of _bt_findinsertloc(), where it remains today). The attached
> patch fixes this, and in passing mentions the fact that
> _bt_insertonpg() only performs retail insertions, and specifically
> never inserts high key items.
>
> I don't think it's necessary to add something about negative infinity
> items to the same comment block. While it's true that _bt_insertonpg()
> cannot truncate downlinks to make new minus infinity items, I see that
> as a narrower issue.
>
--
Teodor Sigaev E-mail: teodor@sigaev.ru
WWW: http://www.sigaev.ru/
Commits
-
Adjust INCLUDE index truncation comments and code.
- 075aade4361b 11.0 landed
-
Add commentary explaining why MaxIndexTuplesPerPage calculation is safe.
- 2a67d6440db4 11.0 cited
-
Indexes with INCLUDE columns and their support in B-tree
- 8224de4f42cc 11.0 landed
-
Add amcheck verification of heap relations belonging to btree indexes.
- 7f563c09f890 11.0 cited
-
Doc: move info for btree opclass implementors into main documentation.
- 3785f7eee3d9 11.0 cited
-
Doc: mention that you can't PREPARE TRANSACTION after NOTIFY.
- e4fbf22831c2 11.0 cited
-
Remove dedicated B-tree root-split record types.
- 0c504a80cf2e 11.0 cited
-
Restructure index access method API to hide most of it at the C level.
- 65c5fcd353a8 9.6.0 cited
-
Split _bt_insertonpg to two functions.
- bc292937ae6a 8.3.0 cited
-
Major overhaul of btree index code. Eliminate special BTP_CHAIN logic for
- 9e85183bfc31 7.1.1 cited