maria-developers team mailing list archive

[Igor Babaev <igor@xxxxxxxxxxxx>]

Daniel Bartholomew wrote:
> On Fri, 19 Nov 2010 09:04:17 -0800
> Igor Babaev <igor@xxxxxxxxxxxx> wrote:
> 
> Igor>     * MySQL Bug #51242
> Igor> <http://bugs.mysql.com/bug.php?id=51242> /HAVING clause on table
> Igor> join produce incorrect results/
> Igor> 
> Igor> Daniel,
> Igor> 
> Igor> this 'fix' was reverted in MariaDB 5.1.48.
> Igor> Then Monty mistakenly pulled in when merging. A month ago I
> Igor> removed it again in 5.1.
> Igor> How has it come that it's there? Is it there?
> Igor> 
> Igor> The same is applied to
> Igor> 
> Igor>     * MySQL Bug #39653 <http://bugs.mysql.com/bug.php?id=39653>
> Igor>       /find_shortest_key in sql_select.cc does not consider
> Igor> clustered primary keys:/
> Igor> 
> Igor> We did not accept the patch in 5.1.48. Did we provide a different
> Igor> patch?
> Igor> 
> Igor> How did you get the list of bugs in
> Igor> http://kb.askmonty.org/v/mariadb-5151-release-notes after all?
> 
> The list of bugs was gatherered from the bugs mentioned in the
> Changelog: http://kb.askmonty.org/v/mariadb-5151-changelog
> 
> The changelog is basically a list of the commit messages between 5.1.50
> and 5.1.51.
> 
> For mybug#51242 you are correct that the fix for this bug was reverted:
> http://bazaar.launchpad.net/~maria-captains/maria/5.1/revision/2952
> I've removed 51242 from the list on the release notes.
> 
> For mybug#39653, the commit message says that a fix was applied for it:
> http://bazaar.launchpad.net/~maria-captains/maria/5.1/revision/2958
> ... but maybe I'm misreading the commit message. Should I remove it
> from the list of fixed bugs?

It seems to me that SergeiG pushed his own fix for this bug which
has nothing to do with mysql patch. And this is important since
Facebook guys screamed exactly about the mysql fix of this bug.
Meanwhile we do not provide a link to our patch. Rather we give a
reference to their patch. I'm afraid people will be confused.

This is true for others of our fixes that differ from those pushed into
the mysql code base.

Regards,
Igor.
> 
> Thanks!
>