public Graph(WeightedBVGraph graph, String[] names) {
org.apache.log4j.Logger logger =
org.apache.log4j.Logger.getLogger("it.unimi.dsi.webgraph.ImmutableGraph");
logger.setLevel(org.apache.log4j.Level.FATAL);
if (names.length != graph.numNodes())
throw new Error("Problem with the list of names for the nodes in the graph.");
try {
File auxFile = File.createTempFile("graph-maps-" + System.currentTimeMillis(), "aux");
auxFile.deleteOnExit();
RecordManager recMan = RecordManagerFactory.createRecordManager(auxFile.getAbsolutePath());
nodes = recMan.hashMap("nodes");
nodesReverse = recMan.hashMap("nodesReverse");
} catch (IOException ex) {
throw new Error(ex);
}
nodes.clear();
nodesReverse.clear();
Constructor[] cons = WeightedArc.class.getDeclaredConstructors();
for (int i = 0; i < cons.length; i++) cons[i].setAccessible(true);
this.graph = graph;
WeightedArcSet list = new WeightedArcSet();
ArcLabelledNodeIterator it = graph.nodeIterator();
while (it.hasNext()) {
if (commit++ % COMMIT_SIZE == 0) {
commit();
list.commit();
}
Integer aux1 = it.nextInt();
Integer aux2 = null;
ArcLabelledNodeIterator.LabelledArcIterator suc = it.successors();
while ((aux2 = suc.nextInt()) != null && aux2 >= 0 && (aux2 < graph.numNodes()))
try {
WeightedArc arc = (WeightedArc) cons[0].newInstance(aux2, aux1, suc.label().getFloat());
list.add(arc);
this.nodes.put(aux1, names[aux1]);
this.nodes.put(aux2, names[aux2]);
this.nodesReverse.put(names[aux1], aux1);
this.nodesReverse.put(names[aux2], aux2);
} catch (Exception ex) {
throw new Error(ex);
}
}
reverse = new WeightedBVGraph(list.toArray(new WeightedArc[0]));
numArcs = list.size();
iterator = nodeIterator();
try {
File auxFile = File.createTempFile("graph" + System.currentTimeMillis(), "aux");
auxFile.deleteOnExit();
String basename = auxFile.getAbsolutePath();
store(basename);
} catch (IOException ex) {
throw new Error(ex);
}
commit();
}
public Graph(String file) throws IOException {
org.apache.log4j.Logger logger =
org.apache.log4j.Logger.getLogger("it.unimi.dsi.webgraph.ImmutableGraph");
logger.setLevel(org.apache.log4j.Level.FATAL);
try {
File auxFile = File.createTempFile("graph-maps-" + System.currentTimeMillis(), "aux");
auxFile.deleteOnExit();
RecordManager recMan = RecordManagerFactory.createRecordManager(auxFile.getAbsolutePath());
nodes = recMan.hashMap("nodes");
nodesReverse = recMan.hashMap("nodesReverse");
} catch (IOException ex) {
throw new Error(ex);
}
nodes.clear();
nodesReverse.clear();
Constructor[] cons = WeightedArc.class.getDeclaredConstructors();
for (int i = 0; i < cons.length; i++) cons[i].setAccessible(true);
String aux = null;
Float weight = (float) 1.0;
WeightedArcSet list = new WeightedArcSet();
BufferedReader br;
try {
br =
new BufferedReader(new InputStreamReader(new GZIPInputStream(new FileInputStream(file))));
} catch (Exception ex) {
br = new BufferedReader(new FileReader(file));
}
while ((aux = br.readLine()) != null)
try {
if (commit++ % COMMIT_SIZE == 0) {
commit();
list.commit();
}
String parts[] = aux.split("\t");
String l1 = new String(parts[0]);
String l2 = new String(parts[1]);
if (!nodesReverse.containsKey(l1)) {
nodesReverse.put(l1, nodesReverse.size());
nodes.put(nodes.size(), l1);
}
if (!nodesReverse.containsKey(l2)) {
nodesReverse.put(l2, nodesReverse.size());
nodes.put(nodes.size(), l2);
}
if (parts.length == 3) weight = new Float(parts[2]);
list.add(
(WeightedArc) cons[0].newInstance(nodesReverse.get(l1), nodesReverse.get(l2), weight));
} catch (Exception ex) {
throw new Error(ex);
}
this.graph = new WeightedBVGraph(list.toArray(new WeightedArc[0]));
br.close();
list = new WeightedArcSet();
br = new BufferedReader(new FileReader(file));
while ((aux = br.readLine()) != null)
try {
if (commit++ % COMMIT_SIZE == 0) {
commit();
list.commit();
}
String parts[] = aux.split("\t");
String l1 = new String(parts[0]);
String l2 = new String(parts[1]);
if (parts.length == 3) weight = new Float(parts[2]);
list.add(
(WeightedArc) cons[0].newInstance(nodesReverse.get(l2), nodesReverse.get(l1), weight));
} catch (Exception ex) {
throw new Error(ex);
}
br.close();
this.reverse = new WeightedBVGraph(list.toArray(new WeightedArc[0]));
numArcs = list.size();
iterator = nodeIterator();
try {
File auxFile = File.createTempFile("graph" + System.currentTimeMillis(), "aux");
auxFile.deleteOnExit();
String basename = auxFile.getAbsolutePath();
store(basename);
} catch (IOException ex) {
throw new Error(ex);
}
commit();
}
public Graph(BVGraph graph) {
org.apache.log4j.Logger logger =
org.apache.log4j.Logger.getLogger("it.unimi.dsi.webgraph.ImmutableGraph");
logger.setLevel(org.apache.log4j.Level.FATAL);
try {
File auxFile = File.createTempFile("graph-maps-" + System.currentTimeMillis(), "aux");
auxFile.deleteOnExit();
RecordManager recMan = RecordManagerFactory.createRecordManager(auxFile.getAbsolutePath());
nodes = recMan.hashMap("nodes");
nodesReverse = recMan.hashMap("nodesReverse");
} catch (IOException ex) {
throw new Error(ex);
}
nodes.clear();
nodesReverse.clear();
Constructor[] cons = WeightedArc.class.getDeclaredConstructors();
for (int i = 0; i < cons.length; i++) cons[i].setAccessible(true);
Integer aux1 = null;
WeightedArcSet list = new WeightedArcSet();
it.unimi.dsi.webgraph.NodeIterator it = graph.nodeIterator();
while ((aux1 = it.nextInt()) != null) {
LazyIntIterator suc = it.successors();
Integer aux2 = null;
while ((aux2 = suc.nextInt()) != null && aux2 >= 0 && (aux2 < graph.numNodes()))
try {
if (commit++ % COMMIT_SIZE == 0) {
list.commit();
}
list.add((WeightedArc) cons[0].newInstance(aux1, aux2, (float) 1.0));
} catch (Exception ex) {
throw new Error(ex);
}
}
this.graph = new WeightedBVGraph(list.toArray(new WeightedArc[0]));
list = new WeightedArcSet();
it = graph.nodeIterator();
while ((aux1 = it.nextInt()) != null) {
LazyIntIterator suc = it.successors();
Integer aux2 = null;
while ((aux2 = suc.nextInt()) != null && aux2 >= 0 && (aux2 < graph.numNodes()))
try {
if (commit++ % COMMIT_SIZE == 0) {
commit();
list.commit();
}
list.add((WeightedArc) cons[0].newInstance(aux2, aux1, (float) 1.0));
this.nodes.put(aux1, "" + aux1);
this.nodes.put(aux2, "" + aux2);
this.nodesReverse.put("" + aux1, aux1);
this.nodesReverse.put("" + aux2, aux2);
} catch (Exception ex) {
throw new Error(ex);
}
}
this.reverse = new WeightedBVGraph(list.toArray(new WeightedArc[0]));
numArcs = list.size();
iterator = nodeIterator();
try {
File auxFile = File.createTempFile("graph" + System.currentTimeMillis(), "aux");
auxFile.deleteOnExit();
String basename = auxFile.getAbsolutePath();
store(basename);
} catch (IOException ex) {
throw new Error(ex);
}
commit();
}
@Override
protected int execute() throws Exception {
BasicConfigurator.configure();
logger.setLevel(Level.INFO);
final ReferenceSequenceFile ref;
try {
ref = ReferenceSequenceFileFactory.getReferenceSequenceFile(refFile);
} catch (Exception e) {
throw new UserException("Couldn't load provided reference sequence file " + refFile, e);
}
variant = parseVariantList(variant);
Comparator<Pair<Integer, File>> positionComparator = new PositionComparator();
Queue<Pair<Integer, File>> priorityQueue;
if (assumeSorted) priorityQueue = new LinkedList<>();
else priorityQueue = new PriorityQueue<>(10000, positionComparator);
FileType fileType = null;
for (File file : variant) {
// if it returns a valid type, it will be the same for all files
fileType = fileExtensionCheck(file, fileType);
if (fileType == FileType.INVALID) return 1;
if (assumeSorted) {
priorityQueue.add(new Pair<>(0, file));
} else {
if (!file.exists()) {
throw new UserException(String.format("File %s doesn't exist", file.getAbsolutePath()));
}
FeatureReader<VariantContext> reader = getFeatureReader(fileType, file);
Iterator<VariantContext> it = reader.iterator();
if (!it.hasNext()) {
System.err.println(
String.format("File %s is empty. This file will be ignored", file.getAbsolutePath()));
continue;
}
VariantContext vc = it.next();
int firstPosition = vc.getStart();
reader.close();
priorityQueue.add(new Pair<>(firstPosition, file));
}
}
FileOutputStream outputStream = new FileOutputStream(outputFile);
EnumSet<Options> options = EnumSet.of(Options.INDEX_ON_THE_FLY);
IndexCreator idxCreator =
GATKVCFUtils.makeIndexCreator(
variant_index_type, variant_index_parameter, outputFile, ref.getSequenceDictionary());
final VariantContextWriter outputWriter =
VariantContextWriterFactory.create(
outputFile, outputStream, ref.getSequenceDictionary(), idxCreator, options);
boolean firstFile = true;
int count = 0;
while (!priorityQueue.isEmpty()) {
count++;
File file = priorityQueue.remove().getSecond();
if (!file.exists()) {
throw new UserException(String.format("File %s doesn't exist", file.getAbsolutePath()));
}
FeatureReader<VariantContext> reader = getFeatureReader(fileType, file);
if (count % 10 == 0) System.out.print(count);
else System.out.print(".");
if (firstFile) {
VCFHeader header = (VCFHeader) reader.getHeader();
outputWriter.writeHeader(header);
firstFile = false;
}
Iterator<VariantContext> it = reader.iterator();
while (it.hasNext()) {
VariantContext vc = it.next();
outputWriter.add(vc);
}
reader.close();
}
System.out.println();
outputWriter.close();
return 0;
}
