From: Michael S. Kaplan (michka@trigeminal.com)
Date: Wed May 09 2007 - 11:05:50 CDT
From: "John H. Jenkins" <jenkins@apple.com>
> On May 9, 2007, at 2:17 AM, Bob Hallissy wrote:
>
>> I don't know about Java, but I do know about TT/OT fonts: Most contain,
>> in their OS/2 table, the ulUnicodeRange and ulCodePageRange bit vectors
>> that should give you a pretty good hint about what characters you might
>> use to display a sample. See
>> http://www.microsoft.com/typography/otspec/os2.htm
>> for detail.
> My experience has been that the values in these fields, particularly for
> older fonts, are not reliable.
Could it be any other way? Even if I create a font that supports every
Cyrillic script code point in Unicode, there is nothing to stop Unicode from
adding more Cyrillic script characters in later versions.
I'd rather consider *that* part of the problem to be a lack of reliability
in the client expectations than in the feature itself. :-)
But for the Unicode subranges, I see very few fonts that add one Arabic code
point and then claim to support the script. There is almost certainly *some*
support there, so if one wants a quick summary view, it is much easier than
digging in to the font.
If one cares about whether specific characters are included, one can
certainly write code to get that information directly.
MichKa [Microsoft]
Blog: http://blogs.msdn.com/michkap
This archive was generated by hypermail 2.1.5 : Wed May 09 2007 - 11:06:43 CDT