gohufont: 2.0 -> 2.1

[rnhmjoj]

Motivation for this change

#35003

Things done

• Tested using sandboxing (nix.useSandbox on NixOS)
• Built on platform(s)
  • NixOS
  • macOS
  • other Linux distributions
• Tested in X.org and linux framebuffer
• Tested compilation of all pkgs that depend on this change (none)
• Tested execution of all binary files (none)
• Determined the impact on package closure size (by running nix path-info -S before and after)
• Assured whether relevant documentation is up to date
• Fits CONTRIBUTING.md.

---

notes: the 2x scaled fonts can't be properly converted to pcf and entirely to psf. It appears to be a problem with the font itself as two different tools are giving the same result.

[rnhmjoj]

Uhm, this is bad. ofBorg, my computer and a different one produced 3 different hashes.

[rnhmjoj]

Ok, it was gzip fault but it's fixed now.

[xeji]

let's try again
@GrahamcOfBorg build gohufont

[xeji]

hash mismatch on all platforms, which I can reproduce locally for x86_64-linux.

[xeji]

fix hash mismatch

[rnhmjoj]

Argh, it must be still nondeterministic: on my machine I can build it with this hash.
I don't know what else to do.

[xeji]

Maybe there's a way to build entirely from github?

[rnhmjoj]

Not yet, until the issue regarding the font size is fixed.
Anyway I bet it's gzip again. I'll try to not compress the pcf fonts.

[rnhmjoj]

@GrahamcOfBorg build gohufont

[rnhmjoj]

@GrahamcOfBorg build gohufont

[rnhmjoj]

I removed the gzipping part and it seems reproducible now.
I don't understand why: there are many fixed-output derivations using gzip, so it's supposedly deterministic... unless there is a bug triggered by these particular files.
Debian reported this issue in the past but it has been fixed, maybe it's a new incarnation of it.

[rnhmjoj]

ping

[rnhmjoj]

Thank you.

[rnhmjoj]

@7c6f434c can you close #35003?