inter: 3.11 -> 3.12

[dtzWill]

Motivation for this change

https://github.com/rsms/inter/releases/tag/v3.12

Things done

• Tested using sandboxing (nix.useSandbox on NixOS, or option sandbox in nix.conf on non-NixOS linux)
• Built on platform(s)
  • NixOS
  • macOS
  • other Linux distributions
• Tested via one or more NixOS test(s) if existing and applicable for the change (look inside nixos/tests)
• Tested compilation of all pkgs that depend on this change using nix-shell -p nixpkgs-review --run "nixpkgs-review wip"
• Tested execution of all binary files (usually in ./result/bin/)
• Determined the impact on package closure size (by running nix path-info -S before and after)
• Ensured that relevant documentation is up to date
• Fits CONTRIBUTING.md.

[worldofpeace]

Darwin build failed, can we fix this?

[dtzWill]

Thanks for catching this! Looks like they have a few variants to choose from-- Hinted ("for Windows") variants, ... err, here's the listing:

https://gist.github.com/dtzWill/e956d2f197404acb5059e43c1ead3533

Not sure why Mac's unzip is behaving different (!) but glad it did because we should make an intentional choice here :).

Hinted version seems the way to go? But upstream's instructions for Linux are to use unhinted-- I'm thinking user's font settings can choose to ignore hinting information and font size doesn't seem a significant concern (for Desktop use).

Thoughts/input welcome, may want to be more particular than just this choice...

[worldofpeace]

Pantheon wants to use this font and they noticed kerning differences between variants elementary/default-settings#152 (comment). I don't really know much about fonts in macos though.

[worldofpeace]

Thanks for catching this! Looks like they have a few variants to choose from-- Hinted ("for Windows") variants, ... err, here's the listing:

https://gist.github.com/dtzWill/e956d2f197404acb5059e43c1ead3533

Not sure why Mac's unzip is behaving different (!) but glad it did because we should make an intentional choice here :).

Hinted version seems the way to go? But upstream's instructions for Linux are to use unhinted-- I'm thinking user's font settings can choose to ignore hinting information and font size doesn't seem a significant concern (for Desktop use).

Thoughts/input welcome, may want to be more particular than just this choice...

I'd think we'd want to distribute all variants? They could even be in different outputs if desired.

[kyrias]

The hinted TTFs are just run through the FreeType2 ttfautohint, rather than being manually hinted. For Linux I would tend to prefer letting FreeType2 do its autohinting instead, both so that you get the choice yourself, and as you might be running a newer version with improved autohinting.

We could certainly distribute the other variants as well however, since people might indeed have uses for the other variants.

[worldofpeace]

Confirmed darwin failure is a regression https://hydra.nixos.org/job/nixpkgs/nixpkgs-20.03-darwin/inter.x86_64-darwin

[dtzWill]

Actually I think the conflict is due to case-sensitivity (or insensitivity on Darwin)-- looks like the conflict is between Inter-Italic.otf and the variable font file inter-italic.otf.

The hinted files are ttf's so I was too hasty before in suggesting the cause of the problem, sorry :).

Not seeing a clearly best answer, anyone?
Personally I'd prefer to avoid splitting into multiple outputs because that sounds like great way to ensure they aren't used ;).

[kyrias]

There's not really any other option though. If someone wants the variable version they need to explicitly choose which one, as otherwise the different fonts would override eachother.

F.ex. Inter (Variable)/Inter.otf, Inter (Variable, single axis)/Inter-italic.otf, and Inter (Variable, single axis)/Inter-roman.otf just have Inter as the font fullname, meaning that you would not be able to choose between them if all of them would be in the same output. At least Inter V (Variable)/Inter-V.otf has the distinct fullname Inter V though.

Theoretically it might be nice to just have Inter (Variable)/Inter.otf as it should cover everything in Inter/, but there are still widely used software that does not support them well. E.g. chromium does not: https://rsms.me/inter/var-test.html.

It also seems like the multi-axis variable version doesn't work in all applications due to the metadata being wrong as well? rsms/inter#142

So Inter V could probably be included in the regular inter output without causing any problems, which would let people use the variable one if they wish. But it would also make the inter output bigger for everyone who doesn't want to use it.

[kyrias]

I'm still of the opinion that the unhinted version should be the default, with the others being available in separate outputs, as that is the arrangement that will work for the mast usecases.

[stale]

I marked this as stale due to inactivity. → More info

[stale]

I marked this as stale due to inactivity. → More info