How can we make Meld for macOS better?

[artmg]

First of all a very big thank you to @yousseb who has poured so much effort into this repo to give us all a chance of running Kai Willadsen's excellent Meld tool on a totally different OS. I pray for the soul of your brother Osama, Youssef, and for the family he has left in your care.

Like many here, I am grateful for the effort that has gone into developing this software, even though I sometimes get frustrated that it can be hard to keep up with the march of updates to the macOS system we run it on, due to the various ways that Apple attempts to protect its customers. I have started this thread to try and get a shared understanding of where development is at with this repo, and its upstream Also to see how willing community members could contribute to making the best, most stable and most easy-to-maintain way to run Meld on all Macs.

First of all, bearing in mind this repo's README has come straight from the upstream, please make sure you have read yousseb's own project home page at https://yousseb.github.io/meld/ – there you will get an understanding not only of how to best use the software in its current state, but also of how the maintainer sees this project in relation to the upstream. Also see the releases, showing not only that this project is definitely not dead, and that yousseb has done well to keep within just one feature version of the upstream. It also indicates ares where the community might be able to help out to improve packages and/or functionality.

So, where are we now? #143 asks if there is a plan for new version, yet release osx-20 shows that there now clearly is. Brew's reports of discontinuation now seem a little unfair.

Ok, where could things go next? In #149 @stlin proposed, and numerous others +1ed, to convert this project into an organisation and possibly add sponsorship. This could help to keep the current functional level working on macOS for a lot longer. However the upstream project continues to develop, currently 55 commits ahead. And they are well along the way towards an upgrade to GTK4, so the delta will only continue to increase.

The alternative suggestion was to Integrate the macOS port upstream. That would mean taking the smarts from this repo and applying them as a PR to the upstream repo. It would probably mean becoming part of the maintenance team up there, too, as the current devs don't seem to have Mac skills or equipment to develop or test. However, they seemed very open to Mac folk joining them. In the long run, that might make it a better outcome for the users who gave this repo 2k2 stars, as the new features just arrive in the mac build. I'm just not sure how that balances the development effort needed – supporting the mac build up there, compared with keeping up with features down here.

I'd love to hear @yousseb on which direction he would rather move in the longer term. Then perhaps if those who might be interested in participating are in agreement, I'd be happy to begin some minor PRs around the documentation, to make things clearer for users and to try and get through some of the issue backlog. I can also get involved in testing when pre-releases are available. To anyone else, what's your view and how could you help?

[dlitster]

I can help with pre-release testing and other housekeeping. I'm not a GTK developer.
It sounds like Rene from this thread is working on a parallel effort to port this into a build pipeline that supports other Gnome/GTK applications.
There's also an background task to migrate from GTK3 to GTK4 which is being worked on.

[dehesselle]

To anyone else, what's your view and how could you help?

My view:

• The best way to honor the work that has been done here is to upstream it.

My help:

• As of a few minutes ago, upstream has macOS CI and produces an app for Apple Silicon (sorry, no Intel).

[Patronics]

For reference in case anyone else, like me, has been searching for the CI-produced version of meld, and/or more information about upstream releases for MacOS, here is the upstream gitlab issue, and here are some (unofficial) releases produced by @dehesselle that can be used until the next update is released upstream.

[artmg]

As it looks like @yousseb has not managed to return to dedicating his time to supporting this project, so it might be tough to make progress here. As @Patronics has mentioned, there is now a dehesselle-meld brew cask, and it seems that the focus has moved onto upstreaming macos functionality into the upstream https://gitlab.gnome.org/GNOME/meld/ project.

it is probably time to close this this issue, and follow the links above to find the future for meld on macos

Thanks to all who have supported and put effort into this project, which helped macos users appreciate the simple yet effective features of meld