Tim Holy
### Rationale Most people who tune into Discourse know about Revise, but I'm guessing that's
This is work towards supporting redefinition of types. It relies on https://github.com/JuliaLang/julia/pull/22721, so it will only be available on 0.7 and higher. AFAICT the only missing part is to populate...
For JET/SnoopCompile integration, it's likely that the same report will be generated for many different calls. This PR makes it easy to distill the collection down to a unique set...
One of the bigger sources of "noise" I've noticed comes from cases where one is analyzing internal methods designed to work with little help from inference and for which therefore...
I wonder if JET would be a good place to consider a potential analysis for latency optimization. A simple example may illustrate the general concept: ```julia julia> methods(match) # 4...
If you want a real-world test case, I got #104 from attempting to use JET on the PaddedViews.jl `runtest.jl` file. I comment out the ambiguity and doctest portions, and then...
In my lab's organization, we have several packages that lack their own documenter documentation, and so don't have a public/private key/secret pair for documenter. However, we do have an organization-level...
https://github.com/timholy/SnoopCompile.jl/pull/256 seems spurious (I always lock the versions of SnoopCompile & SnoopCompileCore together, and release them together)
This may not need to be changed, but just FYI `LinearInterpolation` is *not* free of extrapolation, it merely defines extrapolation with `Throw()` behavior. And this does not come cheap: ```julia...
Plugging away at #226 has been the occasion for reconsidering the overall structure of the package. I've been slowly evolving towards one significant internal API change (introducing `WeightedIndex`) which hopefully...