/**
* Takes an object which subclasses IChemObject, e.g.Molecule, and will read this (from file,
* database, internet etc). If the specific implementation does not support a specific IChemObject
* it will throw an Exception.
*
* @param object The object that subclasses IChemObject
* @return The IChemObject read
* @exception CDKException
*/
public IChemObject read(IChemObject object) throws CDKException {
if (object instanceof IReaction) {
return (IChemObject) readReaction(object.getBuilder());
} else if (object instanceof IReactionSet) {
IReactionSet reactionSet = object.getBuilder().newReactionSet();
reactionSet.addReaction(readReaction(object.getBuilder()));
return reactionSet;
} else if (object instanceof IChemModel) {
IChemModel model = object.getBuilder().newChemModel();
IReactionSet reactionSet = object.getBuilder().newReactionSet();
reactionSet.addReaction(readReaction(object.getBuilder()));
model.setReactionSet(reactionSet);
return model;
} else if (object instanceof IChemFile) {
IChemFile chemFile = object.getBuilder().newChemFile();
IChemSequence sequence = object.getBuilder().newChemSequence();
sequence.addChemModel((IChemModel) read(object.getBuilder().newChemModel()));
chemFile.addChemSequence(sequence);
return chemFile;
} else {
throw new CDKException(
"Only supported are Reaction and ChemModel, and not "
+ object.getClass().getName()
+ ".");
}
}
private IChemFile readChemFile() throws CDKException {
IChemSequence seq = file.getBuilder().newChemSequence();
IChemModel model = file.getBuilder().newChemModel();
IMoleculeSet containerSet = file.getBuilder().newMoleculeSet();
IMolecule container = file.getBuilder().newMolecule();
int lineNumber = 0;
try {
String line = input.readLine();
while (input.ready() && line != null) {
logger.debug((lineNumber++) + ": ", line);
String command = null;
if (isCommand(line)) {
command = getCommand(line);
int lineCount = getContentLinesCount(line);
if ("ATOMS".equals(command)) {
processAtomsBlock(lineCount, container);
} else if ("BONDS".equals(command)) {
processBondsBlock(lineCount, container);
} else if ("IDENT".equals(command)) {
processIdentBlock(lineCount, container);
} else if ("NAME".equals(command)) {
processNameBlock(lineCount, container);
} else {
// skip lines
logger.warn("Dropping block: ", command);
for (int i = 0; i < lineCount; i++) input.readLine();
}
} else {
logger.warn("Unexpected content at line: ", lineNumber);
}
line = input.readLine();
}
containerSet.addAtomContainer(container);
model.setMoleculeSet(containerSet);
seq.addChemModel(model);
file.addChemSequence(seq);
} catch (Exception exception) {
String message = "Error while parsing CTX file: " + exception.getMessage();
logger.error(message);
logger.debug(exception);
throw new CDKException(message, exception);
}
return file;
}
private IReaction checkForXReactionFile(IChemFile chemFile, int numberOfReactions) {
Assert.assertNotNull(chemFile);
Assert.assertEquals(chemFile.getChemSequenceCount(), 1);
org.openscience.cdk.interfaces.IChemSequence seq = chemFile.getChemSequence(0);
Assert.assertNotNull(seq);
Assert.assertEquals(seq.getChemModelCount(), 1);
org.openscience.cdk.interfaces.IChemModel model = seq.getChemModel(0);
Assert.assertNotNull(model);
IReactionSet reactionSet = model.getReactionSet();
Assert.assertNotNull(reactionSet);
Assert.assertEquals(reactionSet.getReactionCount(), numberOfReactions);
IReaction reaction = null;
for (int i = 0; i < numberOfReactions; i++) {
reaction = reactionSet.getReaction(i);
Assert.assertNotNull(reaction);
}
return reaction;
}
private IMolecule checkForXMoleculeFile(IChemFile chemFile, int numberOfMolecules) {
Assert.assertNotNull(chemFile);
Assert.assertEquals(chemFile.getChemSequenceCount(), 1);
org.openscience.cdk.interfaces.IChemSequence seq = chemFile.getChemSequence(0);
Assert.assertNotNull(seq);
Assert.assertEquals(seq.getChemModelCount(), 1);
org.openscience.cdk.interfaces.IChemModel model = seq.getChemModel(0);
Assert.assertNotNull(model);
org.openscience.cdk.interfaces.IMoleculeSet moleculeSet = model.getMoleculeSet();
Assert.assertNotNull(moleculeSet);
Assert.assertEquals(moleculeSet.getMoleculeCount(), numberOfMolecules);
IMolecule mol = null;
for (int i = 0; i < numberOfMolecules; i++) {
mol = moleculeSet.getMolecule(i);
Assert.assertNotNull(mol);
}
return mol;
}
/**
* Read a ChemFile from a file in MDL SDF format.
*
* @return The ChemFile that was read from the MDL file.
*/
private IChemFile readChemFile(IChemFile chemFile) throws CDKException {
IChemSequence chemSequence = chemFile.getBuilder().newInstance(IChemSequence.class);
IChemModel chemModel = chemFile.getBuilder().newInstance(IChemModel.class);
IAtomContainerSet setOfMolecules = chemFile.getBuilder().newInstance(IAtomContainerSet.class);
IAtomContainer m = readAtomContainer(chemFile.getBuilder().newInstance(IAtomContainer.class));
if (m != null && m instanceof IAtomContainer) {
setOfMolecules.addAtomContainer((IAtomContainer) m);
}
chemModel.setMoleculeSet(setOfMolecules);
chemSequence.addChemModel(chemModel);
setOfMolecules = chemFile.getBuilder().newInstance(IAtomContainerSet.class);
chemModel = chemFile.getBuilder().newInstance(IChemModel.class);
String str;
try {
String line;
while ((line = input.readLine()) != null) {
logger.debug("line: ", line);
// apparently, this is a SDF file, continue with
// reading mol files
str = new String(line);
if (str.equals("$$$$")) {
m = readAtomContainer(chemFile.getBuilder().newInstance(IAtomContainer.class));
if (m != null && m instanceof IAtomContainer) {
setOfMolecules.addAtomContainer((IAtomContainer) m);
chemModel.setMoleculeSet(setOfMolecules);
chemSequence.addChemModel(chemModel);
setOfMolecules = chemFile.getBuilder().newInstance(IAtomContainerSet.class);
chemModel = chemFile.getBuilder().newInstance(IChemModel.class);
}
} else {
// here the stuff between 'M END' and '$$$$'
if (m != null) {
// ok, the first lines should start with '>'
String fieldName = null;
if (str.startsWith("> ")) {
// ok, should extract the field name
str.substring(2); // String content =
int index = str.indexOf("<");
if (index != -1) {
int index2 = str.substring(index).indexOf(">");
if (index2 != -1) {
fieldName = str.substring(index + 1, index + index2);
}
}
}
if (line == null) {
throw new CDKException("Expecting data line here, but found null!");
}
StringBuilder data = new StringBuilder();
int dataLineCount = 0;
boolean lineIsContinued = false;
while ((line = input.readLine()) != null) {
if (line.equals(" ") && dataLineCount == 0) {
// apparently a file can have a field whose value is a single space. Moronic
// we check for it *before* trimming it. ideally we should check for any length
// of whitespace
// In adition some SD files have the blank line after the value line contain
// a space, rather than being a true blank line. So we only store a blank value
// line if it's the first line after the key line
data.append(line);
lineIsContinued = false;
dataLineCount++;
if (!lineIsContinued && dataLineCount > 1)
data.append(System.getProperty("line.separator"));
continue;
}
line = line.trim();
if (line.length() == 0) break;
if (line.equals("$$$$")) {
logger.error("Expecting data line here, but found end of molecule: ", line);
break;
}
logger.debug("data line: ", line);
lineIsContinued = false; // reset property
dataLineCount++;
// preserve newlines, unless the line is exactly 80 chars;
// in that case it is assumed to continue on the next line.
// See MDL documentation.
if (!lineIsContinued && dataLineCount > 1)
data.append(System.getProperty("line.separator"));
// add the data line
data.append(line);
// check if the line will be continued on the next line
if (line.length() == 80) lineIsContinued = true;
}
if (fieldName != null) {
logger.info("fieldName, data: ", fieldName, ", ", data);
m.setProperty(fieldName, data.toString());
}
}
}
}
} catch (CDKException cdkexc) {
throw cdkexc;
} catch (Exception exception) {
String error = "Error while parsing SDF";
logger.error(error);
logger.debug(exception);
throw new CDKException(error, exception);
}
try {
input.close();
} catch (Exception exc) {
String error = "Error while closing file: " + exc.getMessage();
logger.error(error);
throw new CDKException(error, exc);
}
chemFile.addChemSequence(chemSequence);
return chemFile;
}
