Luis Michaelis

> Review in progress, meanwhile just overall comment: auto* sf = vm.find_symbol_by_index(func); > Since now `find_symbol_by_index` can return nullptr this makes old code non-robust and requires null-checking. I'll mark what...

> Should it be stored by value? There is little value of keeping connection to VM, as it will be tear-down once hash-map is filled up. Also it makes `operator[]`...

It is. I can upcast it but there is no way of providing raw bytes to `Pixmap`, right? That would be the most efficient way of doing things.

Hm I guess that works. Seems kinda unsafe though. If the `data` member changes its size, my code will break. It would probably better to have a proper constructor.

Well, yes, that is the case right now. But if you ever decide to change when or how `data` is allocated, doing it like this will cause a bug (probably...

Yeah I think that's not possible. But changing that part of the code style shouldn't be too bad since it mostly just affects lines with closing braces, right? I'm not...

Okay, here's the disassembly: ``` luis:Gothic1/ $ objdump -d --start-address=0x748950 --stop-address=0x748A82 -M intel gothic.exe [6:27:11] gothic.exe: file format pei-i386 Disassembly of section .text: 00748950 : 748950: 6a ff push 0xffffffff...

If you need anything else, just let me know :) I'm a noob when it comes to disassembly/de-compilation and binary analysis so I don't really understand any of this in-depth...

It's quite a large function. Here's a [gist](https://gist.github.com/lmichaelis/d4725c09db9eb52cd48a0943ce294033).

> I've never seen this before; the EXE you sent me actually has debug symbols embedded in it that IDA can read. This makes things a LOT easier! I figured....