Issue with Mark to Mark positioning

FontCreator - Font Editor FontCreator - Support
1

As the title suggests, I am having an issue getting mark-to-mark positioning to work properly.

I use a ligature that does the following conversion: combining mark (virama) + base (consonant) = single mark (subscript consonant). I then use mark to base to position this new mark onto the preceding base.

This works fine, but if I use mark to mark to position a subsequent mark (e.g. a vowel sign) onto the mark derived from the ligature, it is ignored.

I assume this is because the mark that the second mark is trying to be positioned onto was originally a base, but I’m not sure what to do to resolve it.

Here is an image to clarify:

  1. pa: p_base
  2. pai: p_base + ai_sign (mark to base working)
  3. paṃ: p_base + anusvara_sign (mark to base working)
  4. paiṃ: p_base + ai_sign + anusvara_sign (mark to base, mark to mark working)
  5. ppa: p_base + (virama + p_base = p_sign) (ligature, mark to base working)
  6. ppai: p_base + (virama + p_base = p_sign) + ai_sign (ligature, mark to base x 2 working)
  7. pppaiṃ: p_base + (virama + p_base = p_sign) + ai_sign + anusvara_sign (ligature, mark to base x2 working, mark to mark working)
  8. pū: p_base + uu_sign
  9. pūṃ: p_base + uu_sign (mark to base working)
  10. ppū: p_base + (virama + p_base = p_sign) + uu_sign (ligature, mark to base working, mark to mark not working)
  11. ppūṃ: p_base + (virama + p_base = p_sign) + uu_sign + anusvara_sign (ligature, mark to base x2 working, mark to mark not working)

Ｔhe uu_sign in 10 and 11 should be affixed to the bottom right of the subscript p, but when I use mark to mark to position it, it is ignored.

I would appreciate any advice.

2

It might really help if you can post a font project with this isolated issue, so only with a small subset of your actual font, so we can look into the problem.

3

Thank you for your reply. I’ve made a basic font with only four glyphs and just the OpenType features required for this specific issue.

For convenience, these are the characters used:
$11E24: 𑸤 (p_base)
$11E40: 𑹀 (uu_sign)
$11E4F: 𑹏 (virama)

pū: 𑸤𑹀
ppa: 𑸤𑹏𑸤
ppū: 𑸤𑹏𑸤𑹀

Any advice would be appreciated.
tocharian test project.fcp (7.61 KB)

4

It seems to work in a web browser, so it is probably an issue in our shaping engine.

If that is the problem then there is nothing to worry about. You just need to test the font outside FontCreator.

5

Thank you for your response. I will test it out myself then.

Thanks again!