katawa-shoujo-re-engineered: compile rpy files in the build phase

[UlyssesZh]

This significantly speeds up launching.

Things done

• Built on platform:
  • x86_64-linux
  • aarch64-linux
  • x86_64-darwin
  • aarch64-darwin
• Tested, as applicable:
  • NixOS tests in nixos/tests.
  • Package tests at passthru.tests.
  • Tests in lib/tests or pkgs/test for functions and "core" functionality.
• Ran nixpkgs-review on this PR. See nixpkgs-review usage.
• Tested basic functionality of all binary files, usually in ./result/bin/.
• Nixpkgs Release Notes
  • Package update: when the change is major or breaking.
• NixOS Release Notes
  • Module addition: when adding a new NixOS module.
  • Module update: when the change is significant.
• Fits CONTRIBUTING.md, pkgs/README.md, maintainers/README.md and other READMEs.

[quantenzitrone]

oh so renpy depends on python-ecdsa which is a purely educational crypto library that is absolutely not meant for usage anywhere
can't even play games without PermittedInsecurePackages

[quantenzitrone]

ok it has been replaced in unreleased renpy

[quantenzitrone]

code looks good other than the nit
game builds, runs and plays fine

i seem to be having heavy memory leaks during cutscenes (~350MB/s), however this is independent of the game being pre-compiled or not, so not an issue with this pr but probably a renpy or ksre upstream bug

[quantenzitrone]

if separate, that definitely belongs in the patchPhase

Suggested change

	configurePhase = ''
	runHook preConfigure
	substituteInPlace game/config.rpy --replace-fail 0.0.0-localbuild ${finalAttrs.version}
	runHook postConfigure
	'';
	postPatch = ''
	substituteInPlace game/config.rpy --replace-fail 0.0.0-localbuild ${finalAttrs.version}
	'';

[UlyssesZh]

Actually this definitely is the configure phase.

[quantenzitrone]

hm i don't know enough about what the configure phase is supposed to be other than autotools specific running the ./configure script.
so since you're so sure, i'll let it go

[quantenzitrone]

bisecting the memory leak is actually impossible
half the time either renpy is broken or the version of ksre has been deleted

[Eveeifyeve]

The worry that I have about this package is this comment here: #490290 (comment) about a memory leak, but if it's impossible to bisect then I wouldn't bother it in this pr. It would be good to create an issue for that upstream to the katawa-shoujo-re-engineered repo so that way people are aware of a memory leak. Everything else looks good.

[quantenzitrone]

renpy/renpy#6934
i actually managed to bisect by running ks:re from a local checkout and smartly choosing commits to try
d96d13a is the culprit (#393386)