stabilizer VS pubgrub

Yes, just because it's from a team connected with the old Stabilizer tool, it doesn't mean it shares the insights and benefits of that kind of testing. At least, that's my understanding of coz, based on my knowledge of https://github.com/ccurtsinger/stabilizer

Nope, he is thinking of that team's previous work, stabilizer. See this talk for more information on both. It is an excellent talk.

Other researchers have tried building tools to address this issue : stabilizer. Unfortunately, stabilizer is now outdated as it hasn't been upgraded from it's LLVM 3.1 foundation. See the project here : https://github.com/ccurtsinger/stabilizer

I see a lot a value in stabilizer.

Any thoughts on using stabilizer in the future?: https://github.com/ccurtsinger/stabilizer

I initially spend quite some time investigating PubGrub for conda (using the awesome rust implementation) but gave up after some time. The biggest issue was that the primitive used in the pubgrub algorithm is the concept of a range and the mathematical operations you can apply to it. This makes the algorithm very good at reasoning about whether or not a particular spec is worth considering. However, it requires that you can express a set of versions as a continuous range. This is often the case when you talk about things like semver, but in conda there are many more factors to consider, for instance, a spec (matchspec in condas case) might, besides referring to a version also refer to a build string in the form of globs ("*cuda") or build numbers, different platforms, features, hashes and you can even use regular expressions in there. I have not been able to figure out a way to map conda matchspecs into a range primitive that works for pubgrub.

Note that the type system is turing complete, so NP-Hard is not so bad. :-) Also if you are interested in working on a useful NP-Hard problem in rust, we would love help with the PubGrub project!

Thanks! it's still a bit fuzzy for me but I've added an issue for this: https://github.com/pubgrub-rs/pubgrub/issues/101 and when we have the time we'll try to implement it!

I am pleased to announce elm-test-rs 1.0.0, a tests runner for the Elm language, like elm-test, mostly built in Rust. This has been a long road and it also included the implementation of the pubgrub dependency solver in Rust to re-conciliate normal and tests dependencies.

If a major pro of Iai is that it can be run in CI, then a guide to setting that up would be great! I would love to use pubgrubs as a guinea pig, if that helps.