Re: UCA: is there a xml version of allkeys.txt available

Martin and K.,

One thing to note is that in CLDR, the Tamil tailoring by default
applies normalization which would not be desirable as a default case for
performance reasons.
I don't know about the other tailorings.

http://unicode.org/repos/cldr/trunk/common/collation/ta.xml

Steven

On 11/17/2011 11:30 PM, "Martin J. Dürst" wrote:
> Hello Marc, others,
>
> On 2011/11/18 2:11, Mark Davis ☕ wrote:
>> There are no planned changes to the Tamil ordering for Unicode 6.1.
>> However, everyone needs to keep in mind that the DUCET is primarily
>> useful
>> for supplying a base ordering, and is rarely suitable for specific
>> languages without a language tailoring. For the language tailorings, you
>> need to look at DUCET *plus *CLDR.
>
> I very much understand that I can't get French and German and Swedish
> sort order at the same time, to this makes a lot of sense for Latin.
>
> However, I assume that the main language written in Tamil script is
> the Tamil language. So I would very much assume that the base ordering
> for the Tamil script should be so that it works out of the box for the
> Tamil language, without any tailoring.
>
> If I got something wrong, please tell me.
>
> Regards, Martin.
>
>> The language tailorings are at
>> http://unicode.org/repos/cldr/trunk/common/collation/ (also beta for
>> CLDR
>> v21). So for Tamil it is
>> http://unicode.org/repos/cldr/trunk/common/collation/ta.xml
>>
>> BTW, there is a chart for the next DUCET at
>> http://www.unicode.org/charts-6.1.0beta/collation/
>>
>> See also:
>>
>> http://unicode.org/collation/
>> http://unicode.org/faq/collation.html
>> http://unicode.org/collation/ducet-changes.html
>> http://unicode.org/collation/ducet-criteria.html
>>
>> Mark
>> *— Il meglio è l’inimico del bene —*
>> *
>> *
>> *
>> [https://plus.google.com/114199149796022210033]
>> *
>>
>>
>>
>> 2011/11/17 கா. சேது | කා. සේතු | K. Sethu<skhome_at_gmail.com>
>>
>>> On Thu, Nov 17, 2011 at 10:18 AM, Mark Davis ☕<mark_at_macchiato.com>
>>> wrote:
>>>> With Unicode 6.1.0, the DUCET is being changes so that the CLDR
>>>> root no
>>>> longer needs a tailoring, so once that is releases it should be
>>>> what you
>>>> want. A beta version is at
>>>> http://unicode.org/Public/UCA/6.1.0/CollationAuxiliary.zip
>>>> (It will be changed before release, so don't use it in production.)
>>>>
>>>> Mark
>>>
>>> Where can we get to see the changes planned? Because as is, the
>>> collation order in it for Tamil letters require a tailoring.
>>>
>>> ~Sethu
>>>
>>
Received on Fri Nov 18 2011 - 09:14:17 CST