Greg Landrum

@bddap : these kinds of things can be quite tricky to track down, but I would definitely like to try. Can you provide code to reproduce the problem?

The first example no longer generates potential chiral centers: ``` In [2]: smi = "O=S1(=O)CC(n2cccn2)C1" In [3]: mol = Chem.MolFromSmiles(smi) In [4]: Chem.FindMolChiralCenters(mol,useLegacyImplementation=False,includeUnassigned=True) Out[4]: [] In [5]: import rdkit In...

I'm going to close this because the code is behaving as expected. It wouldn't be terrible to create an enhancement request to find a heuristic to allow us to reject...

I can't reproduce this. Can you please share the full output of both of your cmake commands?

hmmm, I don't see anything obvious there other than the fact that you are installing boost with homebrew. I have no idea whether that works or not. You might check...

an aside: did your script used to work with a different RDKit release?

Running on linux (using g++) I see a factor of ~4 speedup on the first case (the one with FindAllPathsOfLengthN) and no change for the second example.

> It looks like the vector wrapped in rdChemReactions isn't in rdBase, adding this adds the speedup If this is true, and it makes sense that it would be, how...

@bp-kelley maybe that can be done in a separate PR? If you remove the notebook from this one, I can go ahead and merge it and we'll have the partial...

My attempt to do an out-of-tree build and run the tests works without problems. Here are the commands I ran: ``` cmake -D RDK_INSTALL_INTREE=OFF -D RDK_INSTALL_STATIC_LIBS=OFF -DCMAKE_INSTALL_PREFIX=/scratch/install -D CMAKE_INSTALL_RPATH=/scratch/install/lib -D...