glyph
stop existing, possibly in favor of briefcase
briefcase issues that would need to be resolved to kill Encrust completely:
- https://github.com/beeware/briefcase/issues/2180
- this is needed to embed Sparkle, but perhaps Encrust could continue to exist as a thing that would automate running the thing to generate appcasts for me, since that is inherently platform-specific and fairly customized for me (Encrust doesn't have any code that would readily generalize to a more sophisticated release workflow yet)?
- https://github.com/glyph/Encrust/issues/10
- I need to do my own investigating here to see if e.g. Pomodouroboros could exclude a few directories and get within throwing distance of its current bundle size
Already Completed
- https://github.com/beeware/briefcase/issues/1212
- https://github.com/beeware/briefcase/issues/1217
- https://github.com/beeware/briefcase/issues/1218
- https://github.com/beeware/briefcase/issues/1201
- https://github.com/beeware/briefcase/issues/1221
- https://github.com/beeware/briefcase/issues/1291 — specifically the bit about custom Info.plist keys; right now both of my apps need at least LSUIElement, but likely will need more
maybe also need codesigning-without-building-using-briefcase documentation, but I gather this is already at least possible and maybe the docs even exist, need to read up on that before filing anything there
FYI: beeware/briefcase#2110 (currently in review) resolves the last of your blockers. If there's anything else standing in the way of Briefcase adoption, let us know.
At this point I think this is going to start coming down to the features of py2app, rather than encrust itself. Is there a tutorial for anywhere to use briefcase just for codesigning and archiving, on an existing app bundle? It would be easier to adopt it a bit at a time.
Oh wait I guess I can't quite just use it for codesigning, because I would need some way to invoke the code in https://github.com/beeware/briefcase/issues/1217 too, hmm.
Yeah - you could probably trick Briefcase into thinking it had created the app, and so it could then sign it - but I wouldn't give any guarantees that any trick would be reliable in the long term - what you're describing definitely isn't an officially supported use case.
@freakboy3742 Trying to describe the major things that I think I'm still getting out of py2app.