Not the best examples since they give different results… but the ultimate goal is to be able to get a meaningful description about the crash regardless of what platform it occurred on. This is so in our analytics we can add the meaningful description to the crash report (instead of just an error code) which makes it quicker and easier to debug when we receive the report.
It’s really a tough one and I do sympathise.
If you look at Tracktion Engine, there’s a class called CrashStackTracer which would give you the backtrace. It could serve as inspiration or guide? Though it would be nice to have something like it broken off TE and put into JUCE…
Unfortunately without debug symbols and a stack back trace, or something like that CrashStackTracer, you won’t be able to get much except a vague native platform error code.
1 Like