The STLWRT pre-release progress tracker

[thesquash]

I recognize that STLWRT is taking a lot longer than planned, so I put this up as an impetus for me to get back to work and code whenever I have a spare minute!

This issue is also to provide periodic status updates on the project. It will be closed (but not deleted) when the first stable release is produced.

The current timeline suggests a working, but testing-status, version of STLWRT, version 0.90.0, will be available on 12/25/2020 or whereabouts.

Feel free to add commentary to this issue if you have any; this in an open-discussion issue!

[thesquash]

Geez, I didn't make good on that prediction.

I won't bore you all with another explanation of mine about how that setback occurred.

...except that I realized there's 450,000 lines of code to inspect and STLWRT-ize!

450,000 lines of code is a lot more than I had bargained for.

Now I am coming up with some good (hopefully...) scripts to help me STLWRT-ize the code, so that we can (hopefully soon -- I pray with you all) have this GTK fork and we won't all be stuck with GTK forever.

BTW, to show that I am sorry for what I have done, I added a "thumbs down" reaction to my own post above.

[newbthenewbd]

I know how these kinds of things may feel like when publicly managing a project. Let me just say - You are doing some truly heroic work for the great benefit of us all. You are under no obligation to do it in the first place, let alone to do it faster than you're able to, even when it feels that way.

I, and all of us waiting for STLWRT, and all those who would very impatiently wait for STLWRT if they knew about it but so far don't, we can only be forever grateful for this absolutely amazing project, no matter whether a stable version comes now, or in a couple months - or years from now, for that matter. Please don't overwork yourself to rush it - slow and steady work will peacefully bring a higher-quality product to the finish line than a burnout hell. Not feeling in the right mood to patch up these files, and acting on this lack of the right feel by reading a book instead or what-have-you is not only fine, it might also make the (however backbreaking) work much better.

Thank You!

[camthesaxman]

@thesquash I am fairly experienced with C programming (though I don't know very much about GTK other than dabbling with it for a few side projects). I have a few questions about your plans going forward:

1. How do you plan to deal with incompatibilities between GTK 2 and 3 (such as functions that take different arguments or behave differently between versions)?
2. As the code does not currently compile, why not start with a working build of GTK 2.24 and gradually add GTK3 compatibility on top so that you always maintain a working build?

[stuaxo]

Just here to day thankyou and yes - this will take longer than planned but this is great work you are doing.

Elsewhere CI was mentioned and automation is very helpful for knowing everything works.

[samuel0l0y]

Just wanted to say that there's a project called gtk-classic (or gtk-mushrooms) https://github.com/lah7/gtk3-classic. It may provide a nice way to convert gnome widgets to regular ones that can then be drawn as GTK2 by STLWRT. You may also want to see gtk3-nocsd https://github.com/PCMan/gtk3-nocsd.