Albeit, right now its only an academic paper, with no binary or source code release, this looks extremely interesting (for those with some academic oriented intrigue):
Quote:
|
NoMoreGotos : Decompilation Using Pattern-Independent Control-Flow Structuring and Semantics-Preserving Transformation
|
Quote:
|
https://net.cs.uni-bonn.de/fileadmin/ag/martini/Staff/yakdan/dream_ndss2015.pdf
|
Quote:
|
Helping Johnny to Analyze Malware:A Usability-Optimized Decompiler and Malware Analysis User Study
|
Quote:
|
http://mattsmith.de/pdfs/dream_oakland.pdf
|
The structuring techniques to eliminate gotos without code duplication would enable a very usable and easy to read output and a powerful decompiler. It almost sounds it could decompile code into a more agreeable format than programmers may have originally written it in. Of course goto-free and structuring style is also a matter of preference, the goal of decompilation is to make the most "ease" of reading the code, hence context free grammars of languages and such would be followed as closely as possible. Although subjective, and a few settings might be needed, it seems most of us would have an intuition on looking at the same snippet several times as to which is easiest to read.
Regardless, it would be really interesting to see this decompiler in action...