Lily Wang

Sorry for the delay -- thanks for fixing this @HeetVekariya! Could you please update the CHANGELOG?

Thanks @HeetVekariya -- merging now!

@tylerjereddy I agree with you that there's a lot of downsides to splitting out code from the main library, and the practical outcome in this case seems to be generally...

Thanks @tylerjereddy! I'll start refactoring this unless anyone else would be very keen to do so.

I had interpreted "move StreamLines to analysis" as also refactoring streamlines to be AnalysisBased. It could certainly be moved without refactoring, but if refactoring it's somewhat more work to keep...

Would an argument like `return_self` help here, where the default is the current default behaviour? iirc there were inconsistencies with the behaviour of `enumerate_stereoisomers` and `enumerate_protomers`, as detailed in https://github.com/openforcefield/openff-toolkit/issues/738.

I'm also not incredibly familiar with licenses, but very simplistically the way I understand it is code that runs `import MDAnalysis` should also currently be GPL-licensed. So as @peastman says...

> I don't recall if cosmetic attributes are meant to be serialized out - the docstring of ForceField.to_file implies they are by default. Yes, having ``discard_cosmetic_attributes=False`` by default implies to...

There's #1637, but #1641 is still an issue.

Thanks for the suggestion! I did just comment out that block for my own use -- I've also written my own submission script that loads up the necessary environment for...