rdScaffoldNetwork causes segmenation fault upon None molecule

[AnsgarSchuffenhauer]

Configuration:

• RDKit Version: 2020.03.1
• Operating system: linux (CentOS7)
• Python version (if relevant): 3.8
• Are you using conda? yes
• If you are using conda, which channel did you install the rdkit from?
• If you are not using conda: how did you install the RDKit?

Description:
When feeding a None molecule to rdScaffoldNetwork.CreateScaffoldNetwork() this systemtcially ends up in asegmenation fault. Other regular molecules are processed OK.

>>> from rdkit.Chem.Scaffolds import rdScaffoldNetwork
params = rdScaffoldNetwork.ScaffoldNetworkParams()
params.flattenIsotopes=True
params.includeGenericBondScaffolds=False
params.includeGenericScaffolds=False
params.includeScaffoldsWithAttachments = False
params.includeScaffoldsWithoutAttachments = True
params.pruneBeforeFragmenting = True
sn = rdScaffoldNetwork.CreateScaffoldNetwork([None],params)

==> Segmentation fault

I fully appreciate that I should have checked for proper parsing of the molecule record first, with the mol is not None, and that of course a None molecule cannot generate a valid scaffold network. However, still I would have expected to get an exception in this situation (which I could have contained within a try -except clause), and not a segmentation fault right away.

The issue is not burning, as now that I am aware of the problem, I can prevent the feeding of None molecules.

[greglandrum]

Simple reproducible here:

In [2]: from rdkit.Chem.Scaffolds import rdScaffoldNetwork
In [3]: rdScaffoldNetwork.CreateScaffoldNetwork([None],rdScaffoldNetwork.ScaffoldNetworkParams())

crash!

[greglandrum]

Thanks for reporting this Ansgar. The code should not, under any circumstances, seg fault and crash like this.