Lucida Sans Unicode & Hebrew Hyphen Maqaf

Richard Wordingham richard.wordingham at ntlworld.com
Thu May 5 01:53:31 CDT 2022


On Wed, 4 May 2022 22:12:10 +0000
James Kass via Unicode <unicode at corp.unicode.org> wrote:

> https://docs.microsoft.com/en-us/typography/opentype/spec/gdef
> 
> To clarify, in OpenType parlance a "base glyph" is a single character 
> spacing glyph.  A "mark glyph" is a non-spacing combining glyph.  So
> any spacing glyph which is orthographically considered a mark must be 
> defined as a 'base glyph' in a font's GDEF table.

Not necessarily.

Firstly, whether or not mark glyphs' advance widths are set to zero
depends on the rendering engine for the script.  For a new font it is
most portable over time, application and OS to explicitly set the
advance with to zero for mark glyphs. 

Secondly, an advance width may be set by the GPOS table - the dist
feature is the recommended location for this.  A spacing mark may be
classified as a mark glyph to facilitate glyph substitutions and then
have its real advance width set by the dist feature.  There is or was a
difference in the Microsoft and HarfBuzz interpretation and one had to
be careful with the sequencing of the lookups in GPOS to get the same
effect.  Unfortunately, I can't find the documentation of this.

Richard.




More information about the Unicode mailing list