[HarfBuzz] Character mirroring and the OMPL

[Roozbeh Pournader]

I agree with Jonathan here. The only difference right now is of course a
new pair added to Unicode 6.1, U+27CB and U+27CD MATHEMATICAL RAISING
DIAGONAL and MATHEMATICAL FALLING DIAGONAL. HarfBuzz currently would choose
the mirrored pair from the CMAP of the font if there is no rtla/rtlm lookup
in the font, but that is incorrect according to the OpenType spec.


On Sat, Jan 18, 2014 at 5:04 AM, Jonathan Kew <jfkthame at googlemail.com>wrote:

> We've talked about this at some point in the past, and IIRC Behdad was
> reluctant, but I think we should just fix harfbuzz to respect the OMPL,
> exactly as spec'd. Even though it's an ugly solution.
>
> JK
>
>
> On 18/1/14 11:56, Khaled Hosny wrote:
>
>> Hi Behdad,
>>
>> According to the spec[1] character mirroring uses a fixed OpenType
>> Mirroring Pairs List[2] (I guess for stability in case Unicode changes
>> mirroring properties as was done in the past), but HarfBuzz does not use
>> it and uses the Unicode mirrored property instead, is this intentional?
>>
>> Regards,
>> Khaled
>>
>> 1. http://www.microsoft.com/typography/otspec/TTOCHAP1.htm#ltrrtl
>> 2. http://www.microsoft.com/typography/otspec/ompl.txt
>> _______________________________________________
>> HarfBuzz mailing list
>> HarfBuzz at lists.freedesktop.org
>> http://lists.freedesktop.org/mailman/listinfo/harfbuzz
>>
>>
> _______________________________________________
> HarfBuzz mailing list
> HarfBuzz at lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/harfbuzz
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.freedesktop.org/archives/harfbuzz/attachments/20140119/264a9913/attachment-0001.html>