Sebastian Thiel

That's great to hear! Thanks also for showing a screenshot, as it makes clear that the password entry is bound to having a terminal. So even if everything was configured...

Thanks for sharing! The reason is probably some environment variable that is set in the terminal, but not when launched through other means. Other Git clients, like `Fork` in this...

I don't know if that would *always* work as these shells aren't login shells. And even if so, the `gix-command` one acts like Git does, and it just uses `sh`...

No, that's not what Git does. However, I have added a method on `gix-command::Prepare::with_shell_program()` to allow changing the name or path to the shell. That way it's easy to customise...

The linked PR will now enforce using the current users default shell, assuming it is picked up in the `$SHELL` environment variable, which should make signing more likely. And I...

Thanks so much for sharing, @slamp! I think what's happening there is that GPG tries to launch the pinentry program, but when launched from GitButler there is no terminal that...

Thanks for sharing! Was this ever working? And if so, did that manage to popup a password prompt in the UI?

Thanks a lot for reporting, and for not giving up in getting the installation to work! @ndom91 also works on Linux and might have an idea as to what might...

Yes, I think it's great to keep up the explorative approach using `cargo test` and then put that behind an API for abstraction, along with the first simple 'actual' test....

That's great news! > In its current state, the algorithm only diffs against a single parent. That's interesting - I just know that Git usually ignores merge-commits, but I don't...