JChem issue with MacRoman encoding

There are various formats possible in the instantjchem molecule cells, one of them is

 

<?xml version="1.0"?><cml version="ChemAxon file format v5.10.0, generated by v5.11.3">
<MDocument><MChemicalStruct><molecule molID="m1"><propertyList><property dictRef="CdId" title="CdId"><scalar>18</scalar></property><property dictRef="Mol Weight" title="Mol Weight"><scalar>93.12</scalar></property><property dictRef="Formula" title="Formula"><scalar>C6H7N</scalar></property></propertyList><atomArray atomID="a1 a2 a3 a4 a5 a6 a7" elementType="C C C C C C N" x2="31.803101997879107 30.469422876051073 30.469422876051073 33.136781119707145 33.136781119707145 31.80310199787911 31.80310199787911" y2="-2.106169580076351 -1.33616958007635 0.20383041992365047 0.20383041992365047 -1.3361695800763493 0.9738304199236496 2.51383041992365"></atomArray><bondArray><bond atomRefs2="a5 a1" order="2"></bond><bond atomRefs2="a2 a1" order="1"></bond><bond atomRefs2="a3 a2" order="2"></bond><bond atomRefs2="a5 a4" order="1"></bond><bond atomRefs2="a6 a3" order="1"></bond><bond atomRefs2="a4 a6" order="2"></bond><bond atomRefs2="a7 a6" order="1"></bond></bondArray></molecule></MChemicalStruct></MDocument>
</cml>

 

and another one is

 

<?xml version="1.0" encoding="MacRoman"?><cml version="ChemAxon file format v5.10.0, generated by v5.11.3">
<MDocument><MChemicalStruct><molecule molID="m1"><propertyList><property dictRef="CdId" title="CdId"><scalar>18</scalar></property><property dictRef="Mol Weight" title="Mol Weight"><scalar>93.12</scalar></property><property dictRef="Formula" title="Formula"><scalar>C6H7N</scalar></property></propertyList><atomArray atomID="a1 a2 a3 a4 a5 a6 a7" elementType="C C C C C C N" x2="31.803101997879107 30.469422876051073 30.469422876051073 33.136781119707145 33.136781119707145 31.80310199787911 31.80310199787911" y2="-2.106169580076351 -1.33616958007635 0.20383041992365047 0.20383041992365047 -1.3361695800763493 0.9738304199236496 2.51383041992365"></atomArray><bondArray><bond atomRefs2="a5 a1" order="2"></bond><bond atomRefs2="a2 a1" order="1"></bond><bond atomRefs2="a3 a2" order="2"></bond><bond atomRefs2="a5 a4" order="1"></bond><bond atomRefs2="a6 a3" order="1"></bond><bond atomRefs2="a4 a6" order="2"></bond><bond atomRefs2="a7 a6" order="1"></bond></bondArray></molecule></MChemicalStruct></MDocument>
</cml>

 

The only difference is the encoding.
The MacRoman encoding bit chokes the Jchem Search node (and possibly other nodes as well, i have not tested that) with the following rather silly error:

ERROR     JChem Search     Execute failed: Could not read molecule from byte array.

A workaroud i have found is to simply remove the encoding bit:

update TEST_COMPOUNDS set cd_structure = REPLACE(cd_structure, ' encoding="MacRoman"', '')

But that is quite silly as well, i can not expect the chemists to know about encodings. It should not make the JChem node to go cry in the corner.

Hi,

Thank you for your reporting
Is it possible that you post the stacktrace of this error message?

Thank you for your patience and kindly cooperation.

Best,
Taka

The essential part:

 

 

2012-11-15 12:37:14,736 ERROR KNIME-Worker-1 JChem Search : Execute failed: Could not read molecule from byte array.
2012-11-15 12:37:14,737 DEBUG KNIME-Worker-1 JChem Search : Execute failed: Could not read molecule from byte array.
chemaxon.formats.MolFormatException: Could not read molecule from byte array.
        at chemaxon.jchem.db.JChemSearch.getMolecule(JChemSearch.java:6517)
        at chemaxon.jchem.db.JChemSearch.getHitsAsHitDisplayTool(JChemSearch.java:3671)
        at chemaxon.jchem.db.JChemSearch.getHitsAsMolecules(JChemSearch.java:3567)
        at jp.co.infocom.cheminfo.jchem.jchemsearch.JChemSearchNodeModel.execute(JChemSearchNodeModel.java:614)
        at org.knime.core.node.NodeModel.executeModel(NodeModel.java:536)
        at org.knime.core.node.Node.invokeNodeModelExecute(Node.java:995)
        at org.knime.core.node.Node.execute(Node.java:889)
        at org.knime.core.node.workflow.SingleNodeContainer.performExecuteNode(SingleNodeContainer.java:894)
        at org.knime.core.node.exec.LocalNodeExecutionJob.mainExecute(LocalNodeExecutionJob.java:100)
        at org.knime.core.node.workflow.NodeExecutionJob.run(NodeExecutionJob.java:166)
        at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:439)
        at java.util.concurrent.FutureTask$Sync.innerRun(FutureTask.java:303)
        at java.util.concurrent.FutureTask.run(FutureTask.java:138)
        at org.knime.core.util.ThreadPool$MyFuture.run(ThreadPool.java:124)
        at org.knime.core.util.ThreadPool$Worker.run(ThreadPool.java:239)
Caused by: chemaxon.formats.MolFormatException: Invalid encoding name "MacRoman".
        at chemaxon.marvin.io.formats.cml.MrvImport.readDocument(MrvImport.java:94)
        at chemaxon.marvin.io.MRecordImporter.readMol(MRecordImporter.java:700)
        at chemaxon.marvin.io.MRecordImporter.readMol(MRecordImporter.java:678)
        at chemaxon.marvin.io.MRecordImporter.readMol0(MRecordImporter.java:593)
        at chemaxon.marvin.io.MRecordImporter.readMol(MRecordImporter.java:509)
        at chemaxon.formats.MolImporter.readMol(MolImporter.java:860)
        at chemaxon.formats.MolImporter.read(MolImporter.java:747)
        at chemaxon.formats.MolImporter.read(MolImporter.java:717)
        at chemaxon.util.MolHandler.importMol(MolHandler.java:654)
        at chemaxon.util.MolHandler.setMolecule(MolHandler.java:172)
        at chemaxon.util.MolHandler.<init>(MolHandler.java:105)
        at chemaxon.jchem.db.JChemSearch.getMolecule(JChemSearch.java:6515)
        ... 14 more
Caused by: org.xml.sax.SAXParseException: Invalid encoding name "MacRoman".
        at org.apache.xerces.parsers.DOMParser.parse(Unknown Source)
        at org.apache.xerces.jaxp.DocumentBuilderImpl.parse(Unknown Source)
        at javax.xml.parsers.DocumentBuilder.parse(DocumentBuilder.java:124)
        at chemaxon.marvin.io.formats.cml.MrvImport.readDocument(MrvImport.java:90)
        ... 25 more
2012-11-15 12:37:14,739 DEBUG KNIME-Worker-1 WorkflowManager : JChem Search 0:4 doBeforePostExecution
2012-11-15 12:37:14,739 DEBUG KNIME-Worker-1 NodeContainer : JChem Search 0:4 has new state: POSTEXECUTE
2012-11-15 12:37:14,739 DEBUG KNIME-Worker-1 KnimeResourceNavigator : Node message changed: ERROR: Error in sub flow.
2012-11-15 12:37:14,740 DEBUG KNIME-Worker-1 NodeContainer : MacRoman_error 0 has new state: EXECUTING
2012-11-15 12:37:14,740 DEBUG KNIME-Worker-1 WorkflowManager : JChem Search 0:4 doAfterExecute - failure
2012-11-15 12:37:14,740 DEBUG KNIME-Worker-1 JChem Search : reset
2012-11-15 12:37:14,740 DEBUG KNIME-Worker-1 JChem Search : clean output ports.
2012-11-15 12:37:14,740 DEBUG KNIME-Worker-1 WorkflowFileStoreHandlerRepository : Removing handler c77c616e-40e5-4b51-906e-a1aefa702896 (JChem Search 0:4: <no directory>) - 1 remaining

 

If you want the entire dump i can send it by email.

Any progress yet ?

Hi

I am really sorry I have taken so long to answer your topic.
I have still been investigating this problem, but found a possible solution to the problem.

Hopefully, next version which will be based on JChem 5.12 will fix it.
I apologize for the trouble and thank you for your patience.

Best,
Taka

As a temporary workaround i put the following sql code in an insert and update trigger on the molecules table (mysql)

SET NEW.cd_structure = REPLACE(NEW.cd_structure, ' encoding="MacRoman"', '');

This problem has been fixed in the latest version 2.6.3.v0137 that uses JChem5.12.3.0.
I deeply apologize for my neglect.

Best regards,
Taka