Patrick Thomson
Patrick Thomson
Yup, looks like we need to divest the library core from the `unix` package. This will be feasible once we tackle #136, as the only usages of the `unix` package...
Aha, yeah, we need to qualify the `cabal run` with `exe:semantic` so it doesn’t run the benchmarks. Also, the reason it took an hour is because you compiled with optimizations...
Hmm, interesting. Might be worth trying to upgrade the version of `rules_haskell` that we target; I don’t think Big Sur had come out when we last updated it, and it...
Now that the `tree-sitter-rust` Haskell package exists, next step is to create a `semantic-rust` package like the other ones we have in the semantic repo.
@p-alik Use the `cabal.project` `source-repository-package` feature to pin to the git repo directly.
I tried doing this and observed a significant performance regression with `keras/keras`. This makes me very sad.
Okay I see what the problem is here: `forConcurrently` is not a good solution for us: it does a lot of work per-invocation (creating an `MVar` and calling `fork` for...
We’re currently overhauling how `Semantic.Graph` works, so this code path will probably perish, but let’s keep this open to make sure we account for it during Python stack-graphing.
`forPath` should probably return a `Maybe Language`.
I’ve changed the title of this bug to reflect a specific. To be clear, better PHP support in Semantic doesn’t translate instantly into more GitHub end-user features, as things like...