Dominik Röttsches

Do you mean that the proposed API would only affect step 3? And it would only have an effect if used, otherwise the current behavior as you describe would persist...

> [There was an issue about this but I can't find.] Any chance you meant #1292, but that's not exactly filed with regards to adding API.

Yes, I read #515 again. I am bit a concerned about `hb_buffer_set_not_found_variation_selector_glyph` though in the sense of doing the client having to figure out an unused or suitable glyph id,...

> The rendering code should ignore glyphs having the not-found-variation-selector value. That takes care of last-resort. Do you mean with that that the rendering code should treat them as unshaped...

CC @tursunova

Change in #4529 looks good to me. But I am curious to hear @khaledhosny and @jfkthame's opinion on this, in particular because they were both quite vocal in #515. In...

Thanks for the example. I understand what you meant now. I think technically this code might work for the purpose, but I still see benefit in making that an explicit...

If I read this code right, it asks for U+0020 space glyphs through the glyph lookup callback function. That succeeds when we try the first (from the bottom) from the...

> How would your callback know what to answer? The callback would look at unicode-range of the current subset and only if U+202F would be in the unicode-range of it,...

Works for me if this can be addressed inside of HarfBuzz. Agree that composition/decomposition blurs those lines, too.