Skip to content

Conversation

@KOLANICH
Copy link
Contributor

@KOLANICH KOLANICH commented May 30, 2017

performance issues from #12 should be solved now (use dis.dis(kaitaistruct.KaitaiStream.read_s4be) to understand why).

@GreyCat
Copy link
Member

GreyCat commented May 30, 2017

This is unlikely to be accepted either:

  • This still replaces simple and straightforward code with complex and harder-to-understand one. Basically any code metric (i.e. cyclomatic complexity, maintainability score, etc) indicates that it's more complex that it was.
  • This brings extra dependency.
  • You still haven't provided by benchmarks that support performance claims.
  • Meta-programming machinery still potentially generates problems with using this code with Cython, Python-into-something transpilers, etc, etc.

@KOLANICH KOLANICH closed this May 31, 2017
@KOLANICH
Copy link
Contributor Author

KOLANICH commented May 31, 2017

I have done some benchmarking, it is really slower. Strangely, after using bind it has became even slower (for now I don't know why, on synthetic tests bind gives speed-up, and the byte-code of optimized functions is identical to the original ones).

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants