GithubHelp home page GithubHelp logo

composable-cbdt-poc's People

Contributors

roeln avatar

Watchers

 avatar  avatar  avatar

composable-cbdt-poc's Issues

How to compose large families

Composable Emoji

Target:
Add the ~ 50.000 possible family combinations.

Artwork to compose families from

For each family member, 18 "small" images are needed, totalling 72 images:

Man - Center
Man - Center - Skin tone 1
Man - Center - Skin tone 2
Man - Center - Skin tone 3
Man - Center - Skin tone 4
Man - Center - Skin tone 5

Man - Left
Man - Left - Skin tone 1
Man - Left - Skin tone 2
Man - Left - Skin tone 3
Man - Left - Skin tone 4
Man - Left - Skin tone 5

Man - Right
Man - Right - Skin tone 1
Man - Right - Skin tone 2
Man - Right - Skin tone 3
Man - Right - Skin tone 4
Man - Right - Skin tone 5

Woman - Center
Woman - Center - Skin tone 1
Woman - Center - Skin tone 2
Woman - Center - Skin tone 3
Woman - Center - Skin tone 4
Woman - Center - Skin tone 5

Woman - Left
Woman - Left - Skin tone 1
Woman - Left - Skin tone 2
Woman - Left - Skin tone 3
Woman - Left - Skin tone 4
Woman - Left - Skin tone 5

Woman - Right
Woman - Right - Skin tone 1
Woman - Right - Skin tone 2
Woman - Right - Skin tone 3
Woman - Right - Skin tone 4
Woman - Right - Skin tone 5

Boy - Center
Boy - Center - Skin tone 1
Boy - Center - Skin tone 2
Boy - Center - Skin tone 3
Boy - Center - Skin tone 4
Boy - Center - Skin tone 5

Boy - Left
Boy - Left - Skin tone 1
Boy - Left - Skin tone 2
Boy - Left - Skin tone 3
Boy - Left - Skin tone 4
Boy - Left - Skin tone 5

Boy - Right
Boy - Right - Skin tone 1
Boy - Right - Skin tone 2
Boy - Right - Skin tone 3
Boy - Right - Skin tone 4
Boy - Right - Skin tone 5

Girl - Center
Girl - Center - Skin tone 1
Girl - Center - Skin tone 2
Girl - Center - Skin tone 3
Girl - Center - Skin tone 4
Girl - Center - Skin tone 5

Girl - Left
Girl - Left - Skin tone 1
Girl - Left - Skin tone 2
Girl - Left - Skin tone 3
Girl - Left - Skin tone 4
Girl - Left - Skin tone 5

Girl - Right
Girl - Right - Skin tone 1
Girl - Right - Skin tone 2
Girl - Right - Skin tone 3
Girl - Right - Skin tone 4
Girl - Right - Skin tone 5

Question: do we also want families from gender neutral emoji?

Note: since boy and girl don't change design, we should create left and right versions from the center one by GPOS positioning.

Current situation

Ligature substitutions: one specific combination points to one specific glyph.

Example for man + woman + girl:

<LigatureSet glyph="u1F468">
  ...
  <Ligature components="uni200D,u1F467,uni200D,u1F467" glyph="glyph01760"/>
  ...
</LigatureSet>

These are built based on the filename of the PNG, and aliases defined in a separate text file.

Example of aliases:

The police offer without specified gender gets mapped to police officer + male modifier:

1f46e → 1f46e_200d_2642 # POLICE OFFICER -> man police officer

And police officer without specified gender + light skin tone modifier gets mapped to police officer + light skin tone + male modifier:

1f46e_1f3fb → 1f46e_1f3fb_200d_2642 # light skin tone

Proposed situation

Adding ~ 50.000 ligature lookups isn't going to cut it. We should chain substitutions.

Changes from current process:

  • Can't use direct ligatures
  • We can't build the font based on filenames of the PNG images
    -- Option: generate lookups from generated list

I believe you can generate the lookups from the filenames... https://github.com/davelab6/dancingshoes is really old but might be relevant.

https://github.com/typesupply/feaPyFoFum is a bit newer, same idea.

https://github.com/typesupply/feaTools is also super old and a bit different idea, but might be relevant.

Example replacements

We want a family emoji of:

  • Man (light skin)
  • Woman (medium skin)
  • Girl (dark skin)

So, if + indicates a skin tone modifier, and & a zwj:

man + skin tone 1 & woman + skin tone 3 & girl + skin tone 5

In unicode:

1f468 1f3fb 200d 1f469 1f3fd 200d 1f467 1f3ff

Replacements in this sequence:

Man + skin tone 1    →  Man skin tone 1
Woman + skin tone 3  →  Woman skin tone 3
Girl + skin tone 5   →  Girl skin tone 5

Man skin tone 1 [followed by] zwj + Woman skin tone 3  →  Small man skin tone 1 left
zwj + Woman skin tone 3 [preceded by] Small man skin tone 1 left  →  Small woman skin tone 3 right

zwj + Girl skin tone 5 [preceded by] Small woman skin tone 3 right  →  Small girl skin tone 5

Concerns/questions/notes

Would AFDKO be better at generating complex substitutions? Thus would require significant change in current ttx/FontTools tooling. fontmake can build a UFO with a features.fea file, same as AFDKO.

License etc

Please make a PR to add an OFL license, and an AUTHORs and CONTRIBUTORS file, and assign to me to review.

Author is Google, contributors are you and kabisa. Copyright notice is "the project authors".

FontBakery repo is a good model

Finally, transfer the repo to me and I'll move it to the GF org.

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.