To reproduce:

# install homebrew
ruby -e "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/master/install)";
sudo xcodebuild -license;
# install ots
brew tap davelab6/webfonttools; 
brew update; 
brew install ots --HEAD;
# lint the font
ot-sanitise build/Hubballi-Regular.ttf 

Currently this fails with:

ERROR at src/layout.cc:204 (ParseLookupTable)
ERROR: Layout: Lookup flags require GDEF table, but none was found: 8
ERROR at src/layout.cc:1346 (ParseLookupListTable)
ERROR: Layout: Failed to parse lookup 7
ERROR at src/gsub.cc:618 (ots_gsub_parse)
ERROR: GSUB: Failed to parse lookup list table
ERROR at src/ots.cc:669 (ProcessGeneric)
Failed to sanitise file!

This is fatal and means the font won't work in Chrome and Firefox.

Please add a more descriptive description of the design in README :)

• There are many branches in this repo with various work-in-progress. I don't really see what is what, I think only you know, so please merge and delete the unneeded branches into master, so that is clean and up to date, and set the default branch from feedback to master (in https://github.com/erinmclaughlin/Hubballi/settings)

How are you actually building the UFO into binaries atm?

Hi @erinmclaughlin

I know it's about time, but Hubballi is now released on Google Fonts!
We would like to tweet about it soon, do you have one or more images that could fit twitter better than the images in the README.md ?

Example of tweet: https://twitter.com/googlefonts/status/1487038395918565377

We can also add a small text in using the main script of the font like "New font! Hubballi by @ nametag is a friendly monolinear font supporting…"

Would it be envisageable to have a bolder version ? I'm not that familiar with fonts in general so maybe there is a way to build it from the repository ?

Hi, @schriftgestalt and @davelab6 !
Now that this project has been re-built using Glyphs, the virama (halant) issue is fixed, but another problem has arisen.

Kerning between the subscripts is no longer working as expected. I imagine this must be related to the way the program is dealing with the virama issue. Here's an example: https://github.com/erinmclaughlin/Hubballi/blob/master/source/kannada_two_subscript_example.png

I've double-checked the kerning and subscript sidebearings used in some of these differing combinations, and the values are the same in both files: https://github.com/erinmclaughlin/Hubballi/blob/master/source/metricsmachine_comparison.png

Georg, do you have any idea if it's possible to fix this?

Thanks!

Erin

File info, in case you need it:

-- https://github.com/erinmclaughlin/Hubballi/tree/master/source/development%20ufos/Hubballi-Regular-Glyphs-export.ufo This is the .ufo that I was using in MetricsMachine to re-build and double-check kerning in. I used the "Import Metrics" option in Glyphs to transfer this file's kerning info into to the master Glyphs file, https://github.com/erinmclaughlin/Hubballi/blob/master/source/Hubballi-Regular.glyphs

-- https://github.com/erinmclaughlin/Hubballi/tree/master/source/development%20ufos/Hubballi-Regular-Robofontmu.ufo This is the old Robofont-created .ufo that I am using as a baseline for my comparisons, as it worked correctly.

-- https://github.com/erinmclaughlin/Hubballi/tree/master/source/generated%20test%20fonts The fonts I used in the Impallari test page are here.