Re: [Cdk-devel] [cdk-bugs] [cdk:bugs] #1277 Bond order 4 in MDLV2000 'breaks' AtomTyping

[John M. <joh...@gm...>]

Aha!

That's one I couldn't track down!

J

On 7 Nov 2012, at 20:10, Egon Willighagen <eg...@us...> wrote:

> This issue probably also explains this regression in core:
> 
> ------------- Standard Output ---------------
> Could not match atom: 4 in file bug1208740_1.mol
> Could not match atom: 5 in file bug1208740_1.mol
> Could not match atom: 6 in file bug1208740_1.mol
> Could not match atom: 8 in file bug1208740_1.mol
> Could not match atom: 9 in file bug1208740_1.mol
> Could not match atom: 11 in file bug1208740_1.mol
> Could not match atom: 12 in file bug1208740_1.mol
> Could not match atom: 13 in file bug1208740_1.mol
> Could not match atom: 14 in file bug1208740_1.mol
> Could not match atom: 15 in file bug1208740_1.mol
> Could not match atom: 16 in file bug1208740_1.mol
> Could not match atom: 17 in file bug1208740_1.mol
> Could not match atom: 18 in file bug1208740_1.mol
> Could not match atom: 19 in file bug1208740_1.mol
> Could not match atom: 20 in file bug1208740_1.mol
> Could not match atom: 1 in file bug771485-2.mol
> Could not match atom: 2 in file bug771485-2.mol
> Could not match atom: 3 in file bug771485-2.mol
> Could not match atom: 4 in file bug771485-2.mol
> Could not match atom: 5 in file bug771485-2.mol
> Could not match atom: 6 in file bug771485-2.mol
> Could not match atom: 7 in file bug771485-2.mol
> Could not match atom: 8 in file bug771485-2.mol
> Could not match atom: 9 in file bug771485-2.mol
> Could not match atom: 10 in file bug771485-2.mol
> Could not match atom: 1 in file ring_03419.mol
> Could not match atom: 2 in file ring_03419.mol
> Could not match atom: 3 in file ring_03419.mol
> Could not match atom: 4 in file ring_03419.mol
> Could not match atom: 5 in file ring_03419.mol
> Could not match atom: 6 in file ring_03419.mol
> Could not match atom: 7 in file ring_03419.mol
> Could not match atom: 8 in file ring_03419.mol
> Could not match atom: 9 in file ring_03419.mol
> Could not match atom: 10 in file ring_03419.mol
> Could not match atom: 11 in file ring_03419.mol
> Could not match atom: 12 in file ring_03419.mol
> Could not match atom: 13 in file ring_03419.mol
> Could not match atom: 14 in file ring_03419.mol
> Could not match atom: 15 in file ring_03419.mol
> Could not match atom: 16 in file ring_03419.mol
> Could not match atom: 17 in file ring_03419.mol
> Could not match atom: 18 in file ring_03419.mol
> Could not match atom: 19 in file ring_03419.mol
> Could not match atom: 20 in file ring_03419.mol
> Could not match atom: 21 in file ring_03419.mol
> Could not match atom: 22 in file ring_03419.mol
> Could not match atom: 23 in file ring_03419.mol
> Could not match atom: 24 in file ring_03419.mol
> Could not match atom: 25 in file ring_03419.mol
> Could not match atom: 27 in file ring_03419.mol
> Could not match atom: 28 in file ring_03419.mol
> Could not match atom: 29 in file ring_03419.mol
> Could not match atom: 31 in file ring_03419.mol
> Could not match atom: 32 in file ring_03419.mol
> 
> ------------- Standard Error -----------------
> Fatal Error :1:1: Premature end of file.
> 
> Testcase: testMDLMolfiles(org.openscience.cdk.atomtype.CDKAtomTypeMatcherTestFileReposTest): FAILED
> Could not match all atom types! expected:<2011> but was:<1956>
> junit.framework.AssertionFailedError: Could not match all atom types! expected:<2011> but was:<1956>
> at org.openscience.cdk.atomtype.CDKAtomTypeMatcherTestFileReposTest.testMDLMolfiles(CDKAtomTypeMatcherTestFileReposTest.java:220)
> 
> bugs:1277 Bond order 4 in MDLV2000 'breaks' AtomTyping
> 
> Status: open Created: Wed Nov 07, 2012 09:15 AM UTC by SBeisken Last Updated: Wed Nov 07, 2012 10:16 AM UTC Owner: nobody
> 
> Currently the MDLV2000Reader uses IBond.Order.UNSET for bonds of order 4. This renders atom typing for the atoms of the UNSET bonds impossible because the maximum bond order will always exceed the theoretical pi bond count of the atom type (CDKAtomTypeMatcher 2544). This is a result of the ordinal() method used (container.getMaximumBondOrder(atom).ordinal()). The ordinal of UNSET is 4, greater than most pi bond counts of most common atom types.
> 
> In turn that means that all subsequent methods which rely on typed atoms fail, such as the FixBondOrderTool (which could be used to get around the issue of bond order 4).
> 
> One way around the problem is not to use ordinal. John has posted an alternative approach for 1.4.x (https://sourceforge.net/p/cdk/patches/569/). Adding a custom connections method to the enum would replace ordinal method calls and could return a lower value.
> 
> Sent from sourceforge.net because you indicated interest in https://sourceforge.net/p/cdk/bugs/1277/
> 
> To unsubscribe from further messages, please visit https://sourceforge.net/auth/prefs/
> 
> ------------------------------------------------------------------------------
> LogMeIn Central: Instant, anywhere, Remote PC access and management.
> Stay in control, update software, and manage PCs from one command center
> Diagnose problems and improve visibility into emerging IT issues
> Automate, monitor and manage. Do more in less time with Central
> http://p.sf.net/sfu/logmein12331_d2d_______________________________________________
> Cdk-bugs mailing list
> Cdk...@li...
> https://lists.sourceforge.net/lists/listinfo/cdk-bugs