ArrayIndexOutOfBoundsException in findFirst()

User 870ab5b546

12-12-2006 18:45:17

The code:

Code:

studentMolecule = getSearchMolecule(studentXml, true); // hydrogenize

authorMolecule = getSearchMolecule(authorXml, false); // don't hydro.

MolSearch searchObject = setupSearchObject();

try {

print("hasSubstructure: looking for first occurrence of "

+ "author string in student response");

searchResult = searchObject.findFirst();

...

public MolSearch setupSearchObject() {

MolSearch searchObject = new MolSearch();

searchObject.setSearchType(SearchConstants.SUBSTRUCTURE);

searchObject.setQuery(authorMolecule);

searchObject.setTarget(studentMolecule);

return searchObject;

} // setupSearchObject()

...

public Molecule getSearchMolecule(String xml, boolean hydrogenizeIt)

throws MolFormatException {

MolImporter importer = new MolImporter();

Molecule molecule = importer.importMol(xml);

molecule.ungroupSgroups();

molecule.aromatize();

if (hydrogenizeIt)

molecule = ChemUtils.addImplicitHnoClone(molecule);

molecule = MechSet.normalizeRadicals(molecule);

molecule = MolFunctions.replaceHnoClone(molecule);

return molecule;

} // getSearchMolecule(String, boolean)

(See this post for definitions of the methods in getSearchMolecule().) The student molecule:

Code:

<?xml version="1.0" ?>

<MDocument>

<MPolyline id="o1" color="#000000" headLength="0.8" headWidth="0.5">

<MPoint x="-1.6843750476837158" y="2.40625" />

<MRectanglePoint pos="7" rectRef="o2" />

</MPolyline>

<MRectangle id="o2" color="#000000">

<MPoint x="0.38499999046325684" y="5.149374961853027" />

<MPoint x="9.143750190734863" y="5.149374961853027" />

<MPoint x="9.143750190734863" y="-0.24062499403953552" />

<MPoint x="0.38499999046325684" y="-0.24062499403953552" />

</MRectangle>

<MRectangle id="o3" color="#000000">

<MPoint x="-9.7212495803833" y="5.245625019073486" />

<MPoint x="-1.6843750476837158" y="5.245625019073486" />

<MPoint x="-1.6843750476837158" y="-0.2887499928474426" />

<MPoint x="-9.7212495803833" y="-0.2887499928474426" />

</MRectangle>

<MEFlow id="o4" color="#000000" arcAngle="-150.0" headSkip="0.15"

headLength="0.5" headWidth="0.4" tailSkip="0.15">

<MAtomSetPoint atomRefs="m1.a4 m1.a7" />

<MAtomSetPoint atomRefs="m1.a7 m1.a2" weights="0.25 0.75" />

</MEFlow>

<MChemicalStruct>

<molecule molID="m1">

<atomArray

atomID="a1 a2 a3 a4 a5 a6 a7 a8 a9 a10 a11 a12 a13 a14 a15 a16 a17 a18 a19 a20 a21 a22 a23 a24 a25 a26 a27 a28 a29"

elementType="C C C C C C C C C C C C C C H H H H H H H H H H H H H H H"

formalCharge="0 1 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0"

x2="-8.036875247955322 -6.7031961261272865 -5.369517004299251 -4.035837882471215 -2.7021587606431794 -4.035837882471215 -4.035837882471215 1.8768748283386234 3.210553950166659 4.544233071994695 5.877912193822731 7.211591315650766 5.877912193822731 3.210553950166659 -8.806875247955322 -6.703196126127287 -5.369517004299251 -3.472158760643179 -2.495837882471215 -4.035837882471215 -2.495837882471215 2.6468748283386234 0.5431957065105879 4.544233071994695 7.981591315650766 8.545270437478802 4.3379121938227305 1.670553950166659 3.210553950166659"

y2="1.9731249809265137 2.7431249809265137 1.9731249809265137 2.7431249809265137 1.9731249809265137 1.2031249809265137 4.283124980926514 2.0693749332427966 2.8393749332427967 2.0693749332427966 2.8393749332427967 2.0693749332427966 1.2993749332427966 4.379374933242797 3.3068041027545494 4.283124980926514 3.5131249809265137 0.639445859098478 1.2031249809265137 -0.3368750190734864 4.283124980926514 0.7356958114147611 1.2993749332427966 0.5293749332427966 3.4030540550708324 1.2993749332427966 1.2993749332427964 4.379374933242797 5.919374933242797"

/>

<bondArray>

<bond atomRefs2="a1 a2" order="1" />

<bond atomRefs2="a2 a3" order="1" />

<bond atomRefs2="a3 a4" order="1" />

<bond atomRefs2="a4 a5" order="1" />

<bond atomRefs2="a4 a6" order="1" />

<bond atomRefs2="a4 a7" order="1" />

<bond atomRefs2="a8 a9" order="1" />

<bond atomRefs2="a9 a10" order="1" />

<bond atomRefs2="a10 a11" order="1" />

<bond atomRefs2="a11 a12" order="1" />

<bond atomRefs2="a11 a13" order="1" />

<bond atomRefs2="a9 a14" order="1" />

<bond atomRefs2="a1 a15" order="1" />

<bond atomRefs2="a2 a16" order="1" />

<bond atomRefs2="a3 a17" order="1" />

<bond atomRefs2="a5 a18" order="1" />

<bond atomRefs2="a6 a19" order="1" />

<bond atomRefs2="a6 a20" order="1" />

<bond atomRefs2="a7 a21" order="1" />

<bond atomRefs2="a8 a22" order="1" />

<bond atomRefs2="a8 a23" order="1" />

<bond atomRefs2="a10 a24" order="1" />

<bond atomRefs2="a12 a25" order="1" />

<bond atomRefs2="a12 a26" order="1" />

<bond atomRefs2="a13 a27" order="1" />

<bond atomRefs2="a14 a28" order="1" />

<bond atomRefs2="a14 a29" order="1" />

</bondArray>

</molecule>

</MChemicalStruct>

</MDocument>

The author molecule:

Code:

<?xml version="1.0" ?>

<MDocument>

<MEFlow id="o1" arcAngle="-150.0" headSkip="0.15" headLength="0.5"

headWidth="0.4" tailSkip="0.15">

<MAtomSetPoint atomRefs="m1.a1 m1.a2" />

<MAtomSetPoint atomRefs="m1.a2 m1.a4" weights="0.25 0.75" />

</MEFlow>

<MChemicalStruct>

<molecule molID="m1">

<atomArray

atomID="a1 a2 a3 a4"

elementType="C C C C"

formalCharge="0 0 0 1"

mrvQueryProps="0 A: 0 0"

x2="-13.766666412353516 -12.226666412353515 -14.058333396911621 -12.518333396911622"

y2="2.3333332538604736 2.3333332538604736 -0.5249999761581421 -0.5249999761581421"

/>

<bondArray>

<bond atomRefs2="a1 a2" order="1" />

<bond atomRefs2="a3 a4" order="1" />

</bondArray>

</molecule>

</MChemicalStruct>

</MDocument>

The log:

Code:

MechSubStructSearch::hasSubstructure: setting up search object

MechSubStructSearch::hasSubstructure: looking for first occurrence of author string in student response

MechSubstructure: exception thrown, msg = -1

java.lang.ArrayIndexOutOfBoundsException: -1

at chemaxon.sss.search.StructureSearch.initMolecule(StructureSearch.java:1873)

at chemaxon.sss.search.StructureSearch.initSearch(StructureSearch.java:1430)

at chemaxon.sss.search.StructureSearch.findFirst0(StructureSearch.java:5163)

at chemaxon.sss.search.StructureSearch.findFirst(StructureSearch.java:5156)

at chemaxon.sss.search.MolSearch.findNextEnumerated(MolSearch.java:1114)

at chemaxon.sss.search.MolSearch.findNextFiltered(MolSearch.java:1004)

at chemaxon.sss.search.MolSearch.findFirst(MolSearch.java:896)

at com.prenhall.epoch.mechanisms.MechSubStructSearch.hasSubstructure(MechSubStructSearch.java:87)

???

ChemAxon a3d59b832c

12-12-2006 22:16:48

Bob,

With all your code I managed to reproduce the exception.

The Molecule object which comes into MolSearch as target is inconsistent.

Bond 1 inside atom 0 is not in the molecule and is referring to an atom also not in the molecule.

I have not investigated your code yet, so I don't know why the molecule becomes inconsistent.

Regards,

Szabolcs

ChemAxon a3d59b832c

14-12-2006 13:25:25

I finished the investigation and found that line

Code:

mol.setNode(i, atom);

in your replaceHnoClone() function causes the problem. (I don't exactly know why, maybe because the input is an RgMolecule which itself is a compound molecule consisting other Molecules. I will have our Marvin team check this.)

Fortunately, this call is not necessary, because the atom in question is already part of the molecule. So, using the following simplified function will solve the problem:

Code:

public static Molecule replaceHnoClone(Molecule mol) {

String hydrogenLabel = "H";

int franciumNum = 87;

for (int i = 0; i < mol.getNodeCount(); i++) {

MolAtom atom = (MolAtom) mol.getNode(i);

if (atom.getSymbol().equalsIgnoreCase(hydrogenLabel)) {

atom.setAtno(franciumNum);

}

} // for each node (atom)

return mol;

} // replaceHnoClone(Molecule)

Best regards,

Szabolcs

User 870ab5b546

14-12-2006 14:45:25

Thanks! I made the change, and everything now works.

Interesting. That line in replaceHnoClone() has been in our code forever, and it has never caused this problem before. Either you introduced a bug in the latest JChem, or you fixed a bug that had previously masked this problem.

Have a great holiday season!

ChemAxon a3d59b832c

14-12-2006 14:49:11

bobgr wrote:

Interesting. That line in replaceHnoClone() has been in our code forever, and it has never caused this problem before. Either you introduced a bug in the latest JChem, or you fixed a bug that had previously masked this problem.

Either way, our guys will check and fix it.

bobgr wrote:

Have a great holiday season!

Thank you! And you too!