Mike Hunhoff

I agree that this would be a useful feature to add. We can probably easily adapt [Ghidra's `Jython` code completion](https://github.com/NationalSecurityAgency/ghidra/tree/da94eb86bd2b89c8b0ab9bd89e9f0dc5a3157055/Ghidra/Features/Python/python-src) to Ghidrathon's REPL. Extending that support to an external IDE...

@mahaloz thank you for mentioning this! We can update the documentation to reflect this limitation and possibly clean up the error thrown by Ghidrathon in case 1. Some additional information...

See #55 for progress.

@akhilguruprasad22 I'd love your help here! I've got a [draft PR](https://github.com/mandiant/dncil/pull/55) open that you can continue to develop. It's almost at the finish line but there is some [unaddressed feedback](https://github.com/mandiant/capa/pull/1326)...

I've assigned this issue to you for now. No pressure, if you decide not to move forward please let me know and I'll remove you 😄

@akhilguruprasad22 thank you for all of your research here. To answer your questions 1. > Should blocks which have no predecessors i.e. instructions following unconditional branches be considered block leaders?...

Thanks for the detailed explanations @cblichmann . I'm not seeing the following in practice: > So in the end, there still is a 1:1 mapping between an instruction and its...

> > Specifically, "the previous instruction doesn't have code flow into the current one" would not record e.g. the address of the first instruction of a basic block that follows...

> Based on what I understand @cblichmann said, I think you've found a bug and we should try to develop a test case. Do you see this behavior in all...

@cblichmann I've opened an internal draft to fix this. Please feel free to assign this issue to me.