Closed cpfeiffer closed 10 months ago
Will you provide a patch? Any idea how to solve that case? As far as I understood, there is no real sync mechanism on calling a BrowserFunction. That means, from the Java perspective the call to a BrowserFunction is triggered, and there is no way to sync the browser call with the Java processing, e.g. to wait that another call is triggered until the previous call is finished. At least I am not aware of such a synchronization.
Any ideas or hints welcome.
TBH, I hoped someone else had an idea about this :joy:
This could also just be a problem on Windows/IE and be an SWT issue. At least, I cannot reproduce it on Linux/WebKit2, but then I also suffer from #434 atm.
Unfortunately I have no idea on this. I think I had this issue when I created the example and that is why the ModifyListener is not directly updating the RichTextEditor. Every idea I had on this did not work (queuing, synchronizing, delay, ...)
Not sure what to do here. If nobody else has an idea on this, I tend to close as WONTFIX. But we can also keep it open if someone comes up with an idea.
The only possible improvement I see is wrapping Browser.evaluate()
with a try-catch and improving the error message. The SWTException
's code
is SWT.ERROR_FAILED_EVALUATE
.
That way, other people wouldn't need to spend time debugging this.
try / catch added, thank you
I've modified the rich text editor sample to call
setText()
twice (when pressing the "Set Text" button). This causes a crash:Here's the change, but to try it, comment out the added
getText()
which leads to a crash on startup (#457). https://github.com/GEBIT/nebula/commit/092e69d16e6eedbbbc62cbf976c0d992f2e836a9