[mythtv-users] MythTV vs XBMC

[Sam Jacobs]

On Sat, May 10, 2014 at 8:06 PM, Richard Hulme <peper03 at yahoo.com> wrote:
>
> On 10/05/14 18:05, Sam Jacobs wrote:
>>
>> On Saturday, 10 May 2014, Paul Harrison <mythtv at sky.com
>> <mailto:mythtv at sky.com>> wrote:
>>
>>     On 10/05/14 02:20, Sam Jacobs wrote:
>>
>>         BSkyB customers would think that—Sky have now even put the
>>         search bar itself right on the home screen, and guess how you
>>         enter text into it… Mobile phones that still have number buttons
>>         still use the method, though. Phones that don’t use the method
>>         don’t have the number keys. Imagine how UTTERLY STUPID the
>>         designers of a phone would have to have been, if that phone had
>>         alpha labels but forced you to use the directional buttons to
>>         enter text! Tell me that would not have frustrated the hell out
>>         of you.
>>
>>
>>     I have a Sky HD box and must admit the search does work well.
>>
>>     What you have got to remember though is we have international users
>>     so accented characters would have to be catered for.
>>
>>
>> I was reading a while ago about a Unicode feature, I think it was called
>> something like lexical equivalence, that I think would be useful here.
>> The effect of it in this context is that a bare e might also match è, é,
>> ê and so on. I'll try to find it all again.
>
>
> Hi,
>
> You're probably thinking of primary, secondary and tertiary levels.
>
> Primary compares the base characters (i.e. ignoring all diacritics) but ignores case (so e, E, é, É, è, È, ê and Ê etc. are considered equal).
>
> Secondary also ignores case but differentiates letters with diacritics.
>
> Tertiary differentiates based on case.
>
> I think it varies a bit depending on the language but that's roughly the idea.  I'm no expert on it but I've come across it before.
>
> See http://userguide.icu-project.org/collation/concepts.
>
> Richard.


Hi Richard,

Sorry I didn't reply sooner—your message went into my spam folder for
some reason (Gmail said that it couldn't verify that it was sent by
yahoo.com, or something).

I can't remember what the feature was called. I remember that there
was an interactive demo on the web page, which was quite fun to play
around with!

In any case, it appears that MythTV (or MySQL's LIKE operator) already
behaves like this:

https://www.dropbox.com/s/v2r3wn2qxyrtu9n/mythfrontend%20search%20cafe.png
https://www.dropbox.com/s/4fgd805ghrir1vh/mythfrontend%20search%20café%20%28with%20accent%29.png

Note how a search for "cafe" will match "café", and vice versa. I
don't know about anyone else, but the way I use MythFrontend probably
at least 80% of the occasions where I want to enter text are for a
search of some kind, where it wouldn't matter whether or not the text
was correctly accented, just as MythTV/MySQL ignores case.

That still leaves things like metadata editing, though…

Thanks,
Sam


-- 
Sam Jacobs on MythTV 0.27, UK Freeview SD & Freesat HD, EIT-only EPG
Mac mini, MBE+FE: OS X Mavericks, bundled FE, macports BE
Acer Aspire L310, SBE: Debian testing, Elgato EyeTV for DTT Stick
(Hauppauge Nova-T USB Stick in disguise!) & DVBSky S960 DVB-S2 USB