Kevin Hartman

> By the way, I made PR #807 to address all of the remaining StyleCop warnings Awesome! I will wait for that to go in before fiddling with any warning...

I'd be interested in working on this one!

I'm rethinking a bit of this PR (specifically the stuff with packing and interning) so I'm going to put this into draft for now. The `BitData` stuff is likely to...

Do we want to resurrect this for 2.0, @jakelishman ?

Thanks for the review, @IvanIsCoding! Regarding the Python GIL acquisition in the `PyMatcher` implementations of `NodeMatcher` and `EdgeMatcher`, I think assuming the GIL is acquired is the best thing we...

Thanks for the additional review @IvanIsCoding! > We pass PyFunction's to Dijkstra. Couldn't we do the same here? Then the calls would be fn.call1((a, b)) instead of requiring the py...

> This seems generally fine to me, though given that control_flow is a part of the Operation, does it make sense just to similarly define this somewhere higher up as...

> Long long term, I think control-flow operations may want to move out of being in a PackedInstruction at all, and be first-class in the data description (especially in DAGCircuit,...

> Another API alternative here is to add an implementation of Default / explicit method to Interned where T: ?Sized + ToOwned, ::Owned: Default that constructs the same struct. For...

What was the reason not to impl `Default` itself for `Interned` with constraints on `T`? Otherwise, looks good.