Tzu-ping Chung

I’m more inclined to think `installer`’s CLI shouldn’t get into alternative schemes at all, and should probably remove `--destdir` altogether. But if y’all do want to do this, it’d be...

The obvious way is to use the Python API I guess (which is the primary intention to the project). I’m not saying a CLI flag as you proposed isn’t useful...

No objections from me. Do you have an example what should be added to the repository?

I was thinking about dropping invalid extras as well, but that would be more involved to implement. Nice to have, but not necessary and don’t need to be done right...

Personally I don’t think it’s the library’s place to provide a human-readable output. `ResolutionImpossible` is intended to be caught and handled by the outer call stack, and therefore should provide...

Okay scratch that, the problem is it takes a loooong time to fetch hashes.

The initial round is implemented partially as a “fail fast” mechanism to report immediately if the user-supplied requirement set already contains conflicts. But the same can be achieved with appropriate...

Same here. I don’t like how Pipenv uses subprocesses to run tests though, but maybe we can do better (with the command modules). There must be a better approach than...

The current CLI design follows (loosely) [Sam Boyer’s map-sync-memo proposal](https://medium.com/@sdboyer/so-you-want-to-write-a-package-manager-4ae9c17d9527#3d9e), but with naming tweaks (to match Pipenv). There are conceptually four representations that need to be kept in sync. They...

Uh, I think there are fundamental problems in the resolving process that needs some rewrite. In short, the resolver is not properly backtracking dependency “contributions” when it removes a previous...