vhea table, sign of descender

Hin-Tak Leung htl10 at users.sourceforge.net
Tue Oct 30 00:49:52 CET 2018

I think one small part of the opentype spec probably needs to be reworded slightly. Recently I have an issue with HarBuzz doing strange values for Microsoft mincho for the descender in vhea.
Looking at ISO/IEC 14496-22:2015, section 5.7.7, in version 1.1 version of the table's "vertTypoDescender" description, it says "... For example, a font with an em of 1000 fUnits will set this field to 500".
Then in the next page, the example is "Minus half the em-square height".
I think in the first description a few "minus" and "-" should be added. Thus the anomaly with Microsoft mincho (its descender field in vhea has a +ve value...)
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.aau.at/pipermail/mpeg-otspec/attachments/20181029/58aa1f6c/attachment.html>

More information about the mpeg-otspec mailing list