Comments on: on primary keys http://dom.as/2010/07/30/on-primary-keys/ Sat, 17 Dec 2011 20:14:59 +0000 hourly 1 http://wordpress.com/ By: Domas Mituzas http://dom.as/2010/07/30/on-primary-keys/#comment-1767 Sun, 22 Aug 2010 05:11:15 +0000 http://mituzas.lt/?p=774#comment-1767 Awesome. Thanks, James, for driving this.

]]> By: James Day http://dom.as/2010/07/30/on-primary-keys/#comment-1766 Fri, 20 Aug 2010 18:46:09 +0000 http://mituzas.lt/?p=774#comment-1766 The 17 August patch for this uses the primary key when there’s both a PK and all column secondary index. Expected to be in 5.1.51, just missed the cut off date for 5.1.50. Subject to change, as usual, until it’s actually shipped. For anyone who’s comfortable building the server this is an easy patch to apply.

]]> By: Davi Arnaut http://dom.as/2010/07/30/on-primary-keys/#comment-1765 Fri, 30 Jul 2010 21:52:54 +0000 http://mituzas.lt/?p=774#comment-1765 As you probably know, but reiterating again, we accept contributions. We already accepted a few contributions from MariaDB engineers. Likewise for reversals.

]]> By: Henrik Ingo http://dom.as/2010/07/30/on-primary-keys/#comment-1764 Fri, 30 Jul 2010 21:48:00 +0000 http://mituzas.lt/?p=774#comment-1764 If only upstream MySQL would take note ;-)

Actually, if Oracle has imposed the same clean-room rules on MySQL as they would for their own engineers, I’m not sure MySQL devs are allowed to look at MariaDB code or changelog. For this one, yes, it is not copyrightable “IPR” that we excluded a patch, but if you look at the adjacent one, they probably are not allowed to look at Igor’s optimization and copy it.

Now, if we only could co-operate with them for real then I wouldn’t be speculating about this in the first place….

]]> By: James Day http://dom.as/2010/07/30/on-primary-keys/#comment-1763 Fri, 30 Jul 2010 20:53:44 +0000 http://mituzas.lt/?p=774#comment-1763 Domas, I’ve filed bug #55656 on this. Please provide feedback there, particularly about the suggested fix.

]]> By: Stewart Smith http://dom.as/2010/07/30/on-primary-keys/#comment-1762 Fri, 30 Jul 2010 18:47:25 +0000 http://mituzas.lt/?p=774#comment-1762 Hey, if disk seeks were free… it *might* be better….

]]> By: Igor Babaev http://dom.as/2010/07/30/on-primary-keys/#comment-1761 Fri, 30 Jul 2010 17:09:46 +0000 http://mituzas.lt/?p=774#comment-1761 To be exact: we did not have it reverted in our releases: the patch for #39653 was not accepted after my review and was removed from our internal merge tree. So MariaDB never contained this patch in any release tree.

]]>