neville wrote: ↑Thu Aug 01, 2024 12:55 am
Of course there is the possibility that you could achieve what you want to achieve by going round things in a different way (and, after all, LC?OXT lends itself to doing most things in several ways.
Regretfully, extremely unlikely in this case, but always something to keep in mind.
What you did not really do in your earlier posting is explain what should be the form and the function of your work.
Well it wasn't a bug report, that will come later if indeed it is a bug. I was just fishing to see if someone might have seen the symptoms (flashing of stale buffered screen images when the IDE stutters to a halt) rather than expecting someone to diagnose the disease. And as to what I am trying to achieve, all will be revealed - in my younger days this project would have been completed yesterday, but give me a few months...
Welcome Neville!
(Neville the Devil, is my pet cat's name, btw)
Good to see another 80s HyperCard user like me!
You're right there is a mountain of threads on these forums. I'll try to briefly catch you and any other recent newcomers up...
------------------------------------------------------------------------------------------
First let me say, there is no team(s) here, there's only a community of xTalk refugees.
There are no expert C++ Software Engineers here (so far), so maybe don't get you hopes too high.
Sept. 1st 2021, LC Ltd. abandoned the open-source community edition.
Seth Morrow, Richmond, and myself and a few other people were all trying to collect all of the Community Installers that had been taken off-line (at the time), which we did and are just about all made available on Archive.org.
Then Seth and I discussed continuing FOSS xTalk, in particular keeping the LC community development effort alive. Seth graciously provided hosting for this forum and we began efforts for de-branding the IDE, which was the basic requirement for continuation. I made a new HC inspired icon (see top of page), and Seth and I began editing the many references to LC out of the Dictionary and IDE UI. That took much longer than I initially expected, over a year (I still occasionally find ones that were missed). A de-branded Dictionary was 'compiled'. When I say de-branded what I mean is most references to LC were replaced with a preferably generic term like 'xTalk script', 'the IDE','Extension Builder', etc.
Seth sort of faded to the background since he was too busy with real-life to work on OXT. So I began trying to fix and improve on the IDE mostly on my own, as I saw fit. There were a few contributions here and there from other community members (such as Terry Little), along the way.
I did make some early (unsuccessful) effort to try to recompile the engine on macOS. I'm no X-code C++ expert and I still do not even have an Apple Silicon Mac to work with so M-series CPU support wasn't going to happen from me. But I continued making non-Engine IDE bugs fixes, and I added a few features as well, like an extension macOS native 'DarkMode' support, HyperTalk 'playSentence' support (I'm a long-time MIDI Music tinkerer), new libraries of SVG Icons, things like that.
Then sometime in 2023, Tom Perry (tPerry2x) joined in making his own fork (OXT Lite) with more minimalist design goals. Around that time some other brilliant bloke named 'Tom' produced a binary patcher that fixed the launch-crashing on macOS Sonoma, however this patch may be the cause of some image corruption that Richmond was experience in standalone macOS app builds, so that's still a problem (although there may be a work-around ) that needs to be fixed in the macOS engine's source code.
Tom's joining in was actually perfect timing, because changes to my RL situation has made it so that I have significantly less time available to work on OXT. Tom really picked up the torch and ran. I've tried unsuccessfully to keep up with rolling his changes and fixes into my fork (which is not actually called 'Heavy' it's 'DPE' for Don't Panic Edition). I'm lagging behind.
Tom began trying to recompile the engines, and then we realized then that the 'Pre-Built' binaries that the build process was looking for were behind a password-wall on LC Ltd's server. Fortunately Mark Wielder was able to provide us with those binaries for building on Linux. Tom was able to get Linux Engine recompiled and removed the last piece of 'branding' which was a license checking stack that was embedded into the engines. I was able to compile a few of those missing dependencies from source, but a short while later we were able to get a hold of the rest of those missing binaries. Very recently Mark Welder has patched his fork of LC Community engine repo to pull in and compile those dependency binaries from their source (up to date ). So progress is (slowly) being made on the engine front.
On the IDE side of things both Tom and I continue work trying to make it better however we can.
That's where we are with the 'LC Community Edition' base part of OpenXTalk....
Despite what some people may think, OpenXTalk.org has always been about xTalk and more specifically open-source xTalk.
The LC CE code base may be the best cross-platform open-source xTalk interpreter that we have available to us, but OpenXTalk is
NOT exclusively about the LC CE. (see the OXT Manifesto).
As such any other xTalk/xCard implementations (commercial or FOSS), such as SuperCard, OpenXION, StackSmith, ViperCard, HyperCard Simulator, etc. are more than welcome to be discussed on these forums.
------------------------------------------------------------------------------------------
As for some of your specific questions/comments...
No changes have been made to the GUI-less server version. With the possible exception of Mark's updated Linux build, It's currently the exactly same as it was in LC CE 9.6.3. Although I don't personally use Server engine for anything, I have no intention of removing it. I'm also very interested in updating the Emscripten Web/HTML5 version of the engine.
LCB (for de-branding purposes I've been calling that XB / eXtension Builder) Widgets and Libraries (specially with foreign function interface) in my opinion is an extremely useful piece of the platform, and I have no intentions of removing it (I can't speak for Tom / OXT Lite). It's not at all mutually exclusive to whether or not we support the server version.
I'm not sure what your Font family problems are but I did do some work an Extension that provides some extra macOS font related capabilities. I'm sure if Tom made any fixes for Windows font problems. I'm mostly a Mac user (for now).