François Bérenger comments

Results 420 comments of


                                            François Bérenger

exhaustive mode

this is only reasonable if there is a rather small number of fragments to choose from

exhaustive mode

there must be an integer parameter for safety to prevent looping; like maximum total number of fragments

exhaustive mode

fragments would need to be canonicalized first: keep only one copy of a given fragment per fragments table

when reading molecules, some option to explicitly allow some repair operation

The problem with sanitize=False then doing the "sanitization" later is that there is much more code involved. Correcting the formal charge of 4-valent N atoms is something pretty common. Common...

when reading molecules, some option to explicitly allow some repair operation

So, to be clear, the solution I propose is that we fix the incriminated molecule (if some option is present), warn about it, then go through the whole sanitization procedure....

when reading molecules, some option to explicitly allow some repair operation

Here is some code that does the correction I have in mind: ```python # N with valence 4 becomes N+ def correct_N_formal_charge(mol): for a in mol.GetAtoms(): if a.GetAtomicNum() == 7...

Are `MinPartialCharge` and `MaxPartialCharge` supposed to be within [-500, 500] ?

I think the goal of this code is to find the min and max partial charges in the molecule. 500 is used instead of +inf; -500 is used instead of...

Are `MinPartialCharge` and `MaxPartialCharge` supposed to be within [-500, 500] ?

@maclandrol maybe this can be closed; unless you still have a problem with it

Are `MinPartialCharge` and `MaxPartialCharge` supposed to be within [-500, 500] ?

Ok, so you did find a numerical bug where this code can return inf or nan.

new file: packages/orf/orf.1.0.1/opam

Sorry, now it does; forgot to push the tags.