madshi.net

It's not really good to have API hooks in dlls which you load and unload manually, because that makes unhooking very tricky. If you call UninjectLibrary(), madCodeHook will do some internal "magic". Basically it will unhook all APIs automatically for you, and it will do that *outside* of D...

Hi there,

I'm sorry to say, but unfortunately atm I don't have plans to support ARM, because it would require a major rewrite and I'm currently busy with other projects.

Please double check that "report resource leaks" and "instantly crash on buffer ..." are unchecked here:

http://help.madshi.net/madExceptSettings1.htm

There's already an API exported by madExcept.pas called "NameThread()":
Internally it calls "SetThreadInfo".

Thanks, I've fixed the missing null termination in my source code, as suggested by you guys.

Unfortunately there's no easy way to access the "inner exception" information. You could wait for the completion of the bug report and then use a text parser to check the callstacks, but that seems very ugly, to be honest. Maybe it would make sense to add a try..except block to catch the &...

You need to add madExcept to your uses class to be able to call HideLeak. Hope that helps?

If ignoring the leak this way doesn't work, you can instead hide it by using the HideLeak overload with a callstack.

It's a complicated topic. Basically the calculation of the stack traces can take some time. But for the user it looks weird if the application doesn't do anything for several seconds. So the exception box is opened right away, even before the stack traces are calculated, and then the stack traces ar...

The "Please wait a moment" box only appears in rare/specific circumstances. When exactly do you get it? For example, it appears if one of your exception handlers accesses "exceptIntf.BugReport" before it's fully composed. Maybe instead of tweaking the font sizes, it would be easi...

Well, at least exception stack traces are working ok. I guess you'll have to disable UPX for leak checking, sadly...

Not really, madExcept hooks deep into the RTL and replaces much of the RTL's internal exception handling. Some things unfortunately get broken this way. TThread.FatalException is one of those things.

Could this be UPX related? Does this also happen without UPX?

Not sure why it doesn't work. Does the callstack work for exceptions, though?

FWIW, leak reporting is considered to be a feature that is used on the development PC, not on the end user PC. So it should be no problem doing leak checking with UPX disabled.

Good catch, thank you!

No, I don't have them.