Better database handling of UTF-8 data / Case-insensitive search

[Mark Kretschmann]

On Thu, May 14, 2009 at 11:14 PM, Jeff Mitchell <mitchell at kde.org> wrote:
> Mark Kretschmann wrote:
>> The first point that Stanislav raised (setting the right encoding when
>> creating an entirely new database) seems pretty straightforward,
>> simple, and not too risky (as I wrote). So I would favour including it
>> in 2.1.1.
>>
>> Converting an existing database (his second suggestion) is probably
>> more risky, so I wouldn't touch that in a point release.
>
> Read my reply on amarok at ...basically, we want to do this for everyone at
> the same time, or it will be much more difficult later.  Also, it
> shouldn't be terribly difficult, just a matter of getting the sql
> commands right (which is testable).
>
> I'll agree with Leo here that if I get it in early for 2.1.1 we can put
> it in there.

Actually, after rethinking the situation, I think it would make more
sense to do this _before_ releasing the next beta. If it is risky in
any way, then a beta is the perfect spot. If something goes wrong, we
would then still have one chance to fix it up.

We wouldn't want to risk regressions in a point release - remember Qt
4.5.1 with its totally regressed SVG renderer. Now that was good fun.

-- 
Mark Kretschmann
Amarok Developer
www.kde.org - amarok.kde.org