[Biblical-languages] Unicode "preferred character" note?

Peter Kirk peterkirk at qaya.org
Mon Nov 24 05:58:17 EST 2003


On 24/11/2003 05:30, biblang at ntresources.com wrote:

> I've received some comments from someone working in polytonic Greek font
> design re. a particular Unicode code point. In reference to U+0387, ano
> teleia, the Unicode charts (v. 4) say that "00B7 is the preferred
> character." I've thought that this meant that 00B7 was to be used 
> *instead
> of* 0387, so I did not include a glyph at 0387 in the Unicode version 
> of my
> Galilee font (though I did at 00B7). My correspondent suggests that this
> means only that 0387 is to *look like* 00B7, and that I should include a
> matching glyph at both code points. Can anyone help me sort this out? 
> Thanks.
>
This would be my take on the subject: Unicode recommends that people use 
00B7 rather than 0387, but also recognises that, for historical reasons, 
some people will in fact use 0387. So there are two characters in 
Unicode. Unicode normalisation, of any kind, maps 0387 to 00B7, but text 
is not always normalised and so 0387 may occur in any text.

Unicode actually mandates that canonically equivalent sequences are 
rendered identically, and 0387 and 00B7 are canonically equivalent. The 
recommended way to do this is for the rendering engine to perform some 
kind of normalisation. But Microsoft's Uniscribe engine does not do 
this; I'm not sure about other engines. So the rendering engine is 
likely to ask the font for glyphs for either character. So the best 
thing to do in a font is to support both characters, with links to a 
single glyph.

A keyboard should generate 00B7.

Similarly for 037E and 003B, 0374 and 02B9, 0343 and 0313. If you 
support 0374 you need to support 02B9 as normalisation will replace the 
former by the latter.

-- 
Peter Kirk
peter at qaya.org (personal)
peterkirk at qaya.org (work)
http://www.qaya.org/




More information about the Biblical-languages mailing list