Is this project still maintained?

[BlueskyFR]

Revive of #131

Last commit was about a year ago and @AGWA seems almost inactive for the last ~8+ years if I am not mistaken, so I am opening this issue to get an official statement from him on the future of this project.

At the time of writing, last release was 3.5 years ago.

Some people may claim that the project has reached its "maturity level", but still e.g. 27+ pull requests are piling up

[alerque]

On the previous iteration of this issue I made an offer to come on as a maintainer and later reiterated it. I linked instead of quoted those here because the votes and replies add some weight to the offer.

I'd still be up for being invited to help maintain this repository in place or better yet, starting an org where this could be migrated to so that I can help lead it as a more community effort. I've done this sort of thing before with other projects and have lots of trust credentials in the Linux FOSS ecosystem. I'd be happy to list some of those if trust is the hold up.

[linearizability]

The encryption provided by git-crypt is of paramount importance to me, and there is currently no viable alternative. Certain issues have indeed impacted my usage and remain unresolved. I sincerely hope that its maintenance can continue.

example https://github.com/AGWA/git-crypt/issues/298

[linearizability]

On the previous iteration of this issue I made an offer to come on as a maintainer and later reiterated it. I linked instead of quoted those here because the votes and replies add some weight to the offer.

I'd still be up for being invited to help maintain this repository in place or better yet, starting an org where this could be migrated to so that I can help lead it as a more community effort. I've done this sort of thing before with other projects and have lots of trust credentials in the Linux FOSS ecosystem. I'd be happy to list some of those if trust is the hold up.

@AGWA
As a regular user of this project, I truly appreciate all the work that has been done so far—and I’m genuinely concerned about its future if active maintenance doesn’t resume.

Having reviewed some of the previous discussions, it seems there’s currently no clear plan to restore active maintenance.

I agree with @alerque suggestion that, at a minimum, bringing on 2–3 active maintainers to share the workload would be a practical step forward. As for migrating the project elsewhere, I remain open-minded but also respect the original author’s wishes and autonomy in that decision.

I sincerely hope this project can return to active maintenance—it remains a valuable resource for me and many other users.

[lemonsaurus]

@AGWA

You need to find active maintainers for this project.

We are going to abandon it now because it is clearly not maintained. There is a PR that has been open for nearly 5 years that fixes a problem with multiple git worktrees, why has this not been reviewed and merged? There are dozens of people saying they need this fixed. It's becoming extremely relevant with the explosion of agentic coding in the past months, and this frankly is just not really acceptable. It's completely fine of course if you don't have time to maintain this, but please take the time to find someone who does to take it over.

I would suggest @alerque or others interested in maintaining this consider forking it and posting back here with a link to the fork, this software deserves to live.

[mischadiehm]

For anyone looking for a build with some of the pending PRs included, we keep a fork at narrowin/git-crypt.

It currently includes:

PR	Description
#311	Fix handling small files
#222	Fix multiple worktrees (use common git dir)
#180	Merge driver for secret files
#210	Don't encrypt empty files
#332	Fix textconv producing empty diffs on Linux

The fork also skips encryption of empty files unconditionally, regardless of key age. This fixes fatal: stash failed during merge/stash in repos with older keys (details).

Pre-built binaries for macOS (Apple Silicon) and Linux (amd64) are on the releases page. We've submitted the individual fixes as upstream PRs as well.

The included fixes are tested together with a smoke test suite on macOS and Linux.

[Clindbergh]

Pre-built binaries for macOS (Apple Silicon) and Linux (amd64) are on the releases page.

Do you plan to have any for windows?

[mischadiehm]

sorry we have no windows systems so cant test and wont build.

[reneleonhardt]

sorry we have no windows systems so cant test and wont build.

Are the CI runners here not enough to test and build Windows binaries?

And virtual machines should be free:
https://techbeams.com/tech/microsoft-new-free-virtual-machines-for-developers/
https://windowsreport.com/microsoft-releases-free-windows-11-virtual-machines-available-now-to-download/

[mischadiehm]

Might be but honestly we don't care about Windows. If you need it add it and submit a PR...

[lalibi]

Might be but honestly we don't care about Windows. If you need it add it and submit a PR...

Hi there @mischadiehm , I forked your repo and noticed there’s already a workflow for building a Windows executable, but I don’t see Windows binaries in the releases. Is there a reason for that?

I tried a test release on my fork, and it produced a Windows .exe that seems fine, so I wanted to check whether there are any known issues with it before using it.