Project status / roadmap about canopy

[mindplay-dk]

Oh, I see you do have benchmarks in there :-)

Maybe consider posting the results on the web-site?

[jcoglan]

@mindplay-dk Apologies for taking a while to respond to this, it's been a busy year.

The project doesn't have a defined roadmap as such, it's a hobby project of mine and I work on it as and when I feel like it. It's picked up again recently because I've been using it a lot in other projects.

So essentially it's "done" in the sense it does everything I want of it right now, but that might change in future as I learn more about parsing.

As for benchmarks, I typically avoid publishing benchmarks even if I use them to check things while I work, because they're never representative. Performance will vary based on your grammars and inputs and people shouldn't be making decisions without comparing different tools on their own workloads, rather than relying on benchmarks from a few random examples.

[mindplay-dk]

If you feel like it's complete and stable, maybe consider tagging a release?

As for benchmarks, yeah, I understand and agree with everything you said, but I don't think there's anything misleading about posting them - I mostly like to see a benchmark to reassure myself that (for example) performance for a given grammer isn't detrementally slower compared with another framework. Performance does matter - in some cases it's even crucial. And this framework appears to perform well. If I were you, I would post the result, along with your disclaimer - that's honest, and at least gives someone an idea of what the performance can/might be like :-)

[jcoglan]

I will be publishing version 0.4 when the current train of work on refactoring the internals and adding a couple new features is one. Thanks a lot for your suggestions on benchmarks, I'll consider it when I'm updating the docs :)