Re: [Cdk-jchempaint] More jcp stuff

[Stefan K. <ste...@eb...>]

On Tuesday 30 September 2008 12:24:13 Egon Willighagen wrote:
> On Tue, Sep 30, 2008 at 11:30 AM, Stefan Kuhn <ste...@eb...> wrote:
> > - There is Report->Atom type in the menu. Of these, MM2 and MMFF94 throw
> > exceptions. Anybody having an idea about these and could have a look
> > what's wrong? "Perceive ckd atom type" does not give an error, but what
> > is supposed to happen? Can the atom types be viewed? (The help doesn't
> > tell anything, another problem).
>
> Still CDK1.0, right? What are the exceptions? CDK 1.0 does not have
yes, 1.0
> the CDK atom types, so not sure what that refers to...
Exception in thread "AWT-EventQueue-0" java.lang.NullPointerException
	at 
org.openscience.cdk.atomtype.MM2AtomTypeMatcher.findMatchingAtomType(MM2AtomTypeMatcher.java:109)
	at 
org.openscience.cdk.applications.jchempaint.action.PerceiveAtomTypesAction.actionPerformed(PerceiveAtomTypesAction.java:77)
	at javax.swing.AbstractButton.fireActionPerformed(AbstractButton.java:1995)
line 109 is int atomChemicalGroupConstant = 
((Integer)atom.getProperty(CDKConstants.CHEMICAL_GROUP_CONSTANT)).intValue();
It looks a bit like the property isn't set, but I have no idea about atom 
types
>
> > - If I import the smiles ccccc, there is CH2 on both ends. Is this ok if
> > alternating double bonds are shown?
>
> Depends on what you think the SMILES standard means... OpenSMILES now
> defaults to saying lower case c are SP2 but may only occur in ring
> systems, making the above SMILES invalid. CDK 1.0 sees lower case
> elements as just SP2, so making the above output reasonable... btw...
> the above molecule should also have a single electron somewhere...
> presumable on the middle carbon, as the two outer bonds are double...
>
can we show/handle singe electrons? Anyway, we can just leav it as it is, I 
think.
> Egon