avatarbr wrote:Oh, I did this. Needed to open the metronome to set the speed (reseted when upgraded) when I was in fullscreen. And open the glitter app again.
So, the right thing is to set everything before start the session, and dont change anything during the tease?
About the vocabulary, I hope the autosave chat get my last session (will check later), so I can see the exact point where this happened, because I never know if its a bug or some error in the script, ex:
- Sometimes I got a "pet Name" instead of the #PetName. Long time ago, I had problems with "," or "." after a vocabulary and need to put a space before to avoid errors. But looks working now.
- The particular case I said last time, was a ")" in the middle of the chat. Could be a @RT fail
Anyway, booth error I know the exact script where happened, and will check again.
Sorry... It's me again, explaining stuff in english... What i meant was: The alternated Codeparts are only in action, when the Main Form is redrawn. Any other code is the same. And since the new code uses only .net managed Code, i can't believe the issue related to that. The difference is, the old code did UI-Stuff twice (.Net does what it has to do, and then we overwote the .Net values) and now .Net does it all on it's own. So it's changing performance - and this could be it. T-AI uses timers and the order of execution is unpredictable. The Performance Change could cause it. 1885 accomplished to compensate this by locking the timers using variables. Since we heavily improved the performance, maybe the values don't match anymore. But as the error is unpredictable, it could be a multithreading issue too. I'm writing Code for more than 10 years now and in lots of different languages and i never had a single thread RaceCondition error. Kinda exciting.
There was a ")" in the Chat?
Right now there is an issue with Filtering lists -> what i'm working on right now. But i had to touch @ImageTag and @ShowTaggedImage and dicided to optimize it, since it's unsecure and unefficient. Then while testing the changes, i got distracted by a Script freeze (Another indicator for a race condition issue). I tried to isolate the error, but had no luck yet. But at least i found out why a link is a link and module a module and so on. As i told before: One day i'll stumble over it...
OxiKlein wrote:
As I understand it, wine does not work with apps written with the .NET framework such as this one, and so it will not work with Tease AI. Mono works with .NET apps that use Windows Forms (like Tease AI), but unfortunately it fails to load the app because it does not yet support the text-to-speech engine used with .NET.
That was the only error it threw at me when I tried to run it with mono so it may be possible though to roll an alternate release without the TTS ability and have it then work with mono (if that is the only issue), but it would require further investigation.
Update: Alright I tried compiling the app without speech synthesis and there are still a whole bunch of other errors. I guess it's not just as easy as just removing the TTS, but I suppose it was foolish for me to think it would be that easy. Was worth a shot I guess ¯\_(ツ)_/¯
Wine supports Win32 executablles and also .Net - as long as it's installed. T-AI is 32bit and should run - But this is only based on the specifications.
TTS is not supported by mono and that's the only thing, which causes trouble? How did you disable the TTS feature? Did you remove it completely or only the execution? I guess you need to remove every Reference to system.Speech.
Could you please try it? If it works, we'll use Compiler directives to disable it easily. We still would need to compile it for each version itself, but then we use simply two different compiling configurations.
Edit: nevermind.
But is it really as bad as this mono testing tool showed?
