Closed jverzani closed 5 months ago
Attention: 2 lines
in your changes are missing coverage. Please review.
Comparison is base (
e54c4ee
) 67.75% compared to head (3c7a753
) 67.78%.
Files | Patch % | Lines |
---|---|---|
src/PyCall.jl | 66.66% | 2 Missing :warning: |
:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.
LGTM. I agree about removing hashsalt
.
I can't remember why I did it this way — may be from quite an old Julia version.
Thanks! Let me know if there is anything else to be done here.
@stevengj would it be possible to merge this? Thanks! - Miles
On discourse (https://discourse.julialang.org/t/sympy-sym-s-as-keys-in-dicts/103995) the issue of using tuples of symbolic keys let to the realization that the 2 argument form of
hash
resolved to a different code path than the 1-argument form. This consolidates the two.If this PR is reasonable, it might also make sense to remove the precompile statement for
hashsalt
and thehashsalt
function.The approach taken in
PythonCall
might provide an alternative. (Base.hash(x::Py, h::UInt) = reinterpret(UInt, Int(pyhash(x))) - 3h
)