public Map<String, Long> getRegionSizes(String tableName) {
Map<String, Long> regions = new HashMap<>();
try {
final Table table = connection.getTable(TableName.valueOf(tableName));
RegionLocator regionLocator = connection.getRegionLocator(table.getName());
List<HRegionLocation> tableRegionInfos = regionLocator.getAllRegionLocations();
Set<byte[]> tableRegions = new TreeSet<byte[]>(Bytes.BYTES_COMPARATOR);
for (HRegionLocation regionInfo : tableRegionInfos) {
tableRegions.add(regionInfo.getRegionInfo().getRegionName());
}
ClusterStatus clusterStatus = connection.getAdmin().getClusterStatus();
Collection<ServerName> servers = clusterStatus.getServers();
final long megaByte = 1024L * 1024L;
for (ServerName serverName : servers) {
ServerLoad serverLoad = clusterStatus.getLoad(serverName);
for (RegionLoad regionLoad : serverLoad.getRegionsLoad().values()) {
byte[] regionId = regionLoad.getName();
if (tableRegions.contains(regionId)) {
long regionSizeBytes = regionLoad.getStorefileSizeMB() * megaByte;
regions.put(regionLoad.getNameAsString(), regionSizeBytes);
}
}
}
} catch (IOException e) {
e.printStackTrace();
}
return regions;
}
@Override
public void readFields(DataInput input) throws IOException {
byte[] tableNameBytes = Bytes.readByteArray(input);
PName tableName = new PNameImpl(tableNameBytes);
PTableType tableType = PTableType.values()[WritableUtils.readVInt(input)];
long sequenceNumber = WritableUtils.readVLong(input);
long timeStamp = input.readLong();
byte[] pkNameBytes = Bytes.readByteArray(input);
String pkName = pkNameBytes.length == 0 ? null : Bytes.toString(pkNameBytes);
int nColumns = WritableUtils.readVInt(input);
List<PColumn> columns = Lists.newArrayListWithExpectedSize(nColumns);
for (int i = 0; i < nColumns; i++) {
PColumn column = new PColumnImpl();
column.readFields(input);
columns.add(column);
}
Map<String, byte[][]> guidePosts = new HashMap<String, byte[][]>();
int size = WritableUtils.readVInt(input);
for (int i = 0; i < size; i++) {
String key = WritableUtils.readString(input);
int valueSize = WritableUtils.readVInt(input);
byte[][] value = new byte[valueSize][];
for (int j = 0; j < valueSize; j++) {
value[j] = Bytes.readByteArray(input);
}
guidePosts.put(key, value);
}
PTableStats stats = new PTableStatsImpl(guidePosts);
init(tableName, tableType, timeStamp, sequenceNumber, pkName, columns, stats);
}
@Override
public PColumnFamily getColumnFamily(String familyName) throws ColumnFamilyNotFoundException {
PColumnFamily family = familyByString.get(familyName);
if (family == null) {
throw new ColumnFamilyNotFoundException(familyName);
}
return family;
}
@Override
public PColumnFamily getColumnFamily(byte[] familyBytes) throws ColumnFamilyNotFoundException {
PColumnFamily family = familyByBytes.get(familyBytes);
if (family == null) {
String familyName = Bytes.toString(familyBytes);
throw new ColumnFamilyNotFoundException(familyName);
}
return family;
}
@Override
public QueryResult<Variant> getAllVariantsByRegionAndStudy(
Region region, String sourceId, QueryOptions options) {
Long start, end, dbstart, dbend;
start = System.currentTimeMillis();
QueryResult<Variant> queryResult =
new QueryResult<>(
String.format("%s:%d-%d", region.getChromosome(), region.getStart(), region.getEnd()));
List<Variant> results = new LinkedList<>();
boolean includeSamples;
boolean includeStats;
boolean includeEffects;
if (!options.containsKey("samples")
&& !options.containsKey("stats")
&& !options.containsKey("effects")) {
includeSamples = true;
includeStats = true;
includeEffects = true;
} else {
includeSamples = options.containsKey("samples") && options.getBoolean("samples");
includeStats = options.containsKey("stats") && options.getBoolean("stats");
includeEffects = options.containsKey("effects") && options.getBoolean("effects");
}
try {
String startRow = buildRowkey(region.getChromosome(), Long.toString(region.getStart()));
String stopRow = buildRowkey(region.getChromosome(), Long.toString(region.getEnd()));
HTable table = new HTable(admin.getConfiguration(), tableName);
dbstart = System.currentTimeMillis();
Scan regionScan = new Scan(startRow.getBytes(), stopRow.getBytes());
ResultScanner scanres = table.getScanner(regionScan);
dbend = System.currentTimeMillis();
queryResult.setDbTime(dbend - dbstart);
// Iterate over results and, optionally, their samples and statistics
for (Result result : scanres) {
String[] rowkeyParts = new String(result.getRow(), CHARSET_UTF_8).split("_");
String chromosome = rowkeyParts[0].replaceFirst("^0+(?!$)", "");
int position = Integer.parseInt(rowkeyParts[1]);
// Get basic result fields from Protocol Buffers message
NavigableMap<byte[], byte[]> infoMap = result.getFamilyMap("i".getBytes());
byte[] byteInfo = infoMap.get((sourceId + "_data").getBytes());
VariantFieldsProtos.VariantInfo protoInfo =
VariantFieldsProtos.VariantInfo.parseFrom(byteInfo);
String reference = protoInfo.getReference();
String alternate = StringUtils.join(protoInfo.getAlternateList(), ",");
String format = StringUtils.join(protoInfo.getFormatList(), ":");
Variant variant = new Variant(chromosome, position, position, reference, alternate);
// Set samples if requested
if (includeSamples) {
NavigableMap<byte[], byte[]> sampleMap = result.getFamilyMap("d".getBytes());
Map<String, Map<String, String>> resultSampleMap = new HashMap<>();
// Set samples
for (byte[] s : sampleMap.keySet()) {
String sampleName = (new String(s, CHARSET_UTF_8)).replaceAll(sourceId + "_", "");
VariantFieldsProtos.VariantSample sample =
VariantFieldsProtos.VariantSample.parseFrom(sampleMap.get(s));
String sample1 = sample.getSample();
String[] values = sample1.split(":");
String[] fields = format.split(":");
Map<String, String> singleSampleMap = new HashMap<>();
for (int i = 0; i < fields.length; i++) {
singleSampleMap.put(fields[i], values[i]);
}
// TODO
// variant.addSampleData(sampleName, singleSampleMap);
}
}
// Set stats if requested
if (includeStats) {
byte[] byteStats = infoMap.get((sourceId + "_stats").getBytes());
VariantFieldsProtos.VariantStats protoStats =
VariantFieldsProtos.VariantStats.parseFrom(byteStats);
VariantStats variantStats =
new VariantStats(
chromosome,
position,
reference,
alternate,
protoStats.getMaf(),
protoStats.getMgf(),
protoStats.getMafAllele(),
protoStats.getMgfGenotype(),
protoStats.getMissingAlleles(),
protoStats.getMissingGenotypes(),
protoStats.getMendelianErrors(),
protoStats.getIsIndel(),
protoStats.getCasesPercentDominant(),
protoStats.getControlsPercentDominant(),
protoStats.getCasesPercentRecessive(),
protoStats.getControlsPercentRecessive());
variant.setStats(variantStats);
}
if (includeEffects) {
QueryResult<VariantEffect> queryEffects = getEffectsByVariant(variant, options);
variant.setEffect(queryEffects.getResult());
}
results.add(variant);
}
} catch (IOException e) {
System.err.println(e.getClass().getName() + ": " + e.getMessage());
}
queryResult.setResult(results);
queryResult.setNumResults(results.size());
end = System.currentTimeMillis();
queryResult.setTime(end - start);
return queryResult;
}
public List<Variant> getRecordSimpleStats(
String study, int missing_gt, float maf, String maf_allele) {
BasicDBObject compare =
new BasicDBObject("studies.stats.allele_maf", maf_allele)
.append("studies.stats.MAF", maf)
.append("studies.stats.missing", missing_gt);
List<Get> hbaseQuery = new ArrayList<>();
DBCollection collection = db.getCollection("variants");
Iterator<DBObject> result = collection.find(compare);
String chromosome = new String();
while (result.hasNext()) {
DBObject variant = result.next();
String position = variant.get("_id").toString();
// hbase query construction
Get get = new Get(position.getBytes());
hbaseQuery.add(get);
}
// Complete results, from HBase
tableName = study;
effectTableName = tableName + "effect";
Map<String, Variant> resultsMap = new HashMap<>();
try {
HTable table = new HTable(admin.getConfiguration(), tableName);
HTable effectTable = new HTable(admin.getConfiguration(), effectTableName);
Result[] hbaseResultEffect = effectTable.get(hbaseQuery);
Result[] hbaseResultStats = table.get(hbaseQuery);
// List<Variant> results = new LinkedList<>();
for (Result r : hbaseResultStats) {
String position = new String(r.getRow(), CHARSET_UTF_8);
String[] aux = position.split("_");
String inner_position = aux[1];
String chr = aux[0];
// position parsing
if (chr.startsWith("0")) {
chr = chr.substring(1);
}
while (inner_position.startsWith("0")) {
inner_position = inner_position.substring(1);
}
List<VariantFieldsProtos.VariantSample> samples = new LinkedList<>();
NavigableMap<byte[], byte[]> infoMap = r.getFamilyMap("i".getBytes());
byte[] byteStats = infoMap.get((study + "_stats").getBytes());
VariantFieldsProtos.VariantStats stats =
VariantFieldsProtos.VariantStats.parseFrom(byteStats);
byte[] byteInfo = infoMap.get((study + "_data").getBytes());
VariantFieldsProtos.VariantInfo info = VariantFieldsProtos.VariantInfo.parseFrom(byteInfo);
String alternate = StringUtils.join(info.getAlternateList(), ", ");
String reference = info.getReference();
Variant partialResult =
new Variant(
chr,
Integer.parseInt(inner_position),
Integer.parseInt(inner_position),
reference,
alternate);
String format = StringUtils.join(info.getFormatList(), ":");
NavigableMap<byte[], byte[]> sampleMap = r.getFamilyMap("d".getBytes());
Map<String, Map<String, String>> resultSampleMap = new HashMap<>();
// StringBuilder sampleRaw = new StringBuilder();
for (byte[] s : sampleMap.keySet()) {
String qual = (new String(s, CHARSET_UTF_8)).replaceAll(study + "_", "");
VariantFieldsProtos.VariantSample sample =
VariantFieldsProtos.VariantSample.parseFrom(sampleMap.get(s));
String sample1 = sample.getSample();
String[] values = sample1.split(":");
String[] fields = format.split(":");
Map<String, String> singleSampleMap = new HashMap<>();
for (int i = 0; i < fields.length; i++) {
singleSampleMap.put(fields[i], values[i]);
}
resultSampleMap.put(qual, singleSampleMap);
}
VariantStats variantStats =
new VariantStats(
chromosome,
Integer.parseInt(inner_position),
reference,
alternate,
stats.getMaf(),
stats.getMgf(),
stats.getMafAllele(),
stats.getMgfGenotype(),
stats.getMissingAlleles(),
stats.getMissingGenotypes(),
stats.getMendelianErrors(),
stats.getIsIndel(),
stats.getCasesPercentDominant(),
stats.getControlsPercentDominant(),
stats.getCasesPercentRecessive(),
stats.getControlsPercentRecessive());
partialResult.setStats(variantStats);
resultsMap.put(new String(r.getRow(), CHARSET_UTF_8), partialResult);
}
for (Result r : hbaseResultEffect) {
if (!r.isEmpty()) {
NavigableMap<byte[], byte[]> effectMap = r.getFamilyMap("e".getBytes());
Variant partialResult = resultsMap.get(new String(r.getRow(), CHARSET_UTF_8));
System.out.println("Recuperado " + partialResult.toString());
String s = partialResult.getReference() + "_" + partialResult.getAlternate();
VariantEffectProtos.EffectInfo effectInfo =
VariantEffectProtos.EffectInfo.parseFrom(effectMap.get(s.getBytes()));
VariantEffect variantEffect =
new VariantEffect(
partialResult.getChromosome(),
(int) partialResult.getStart(),
partialResult.getReference(),
partialResult.getAlternate(),
effectInfo.getFeatureId(),
effectInfo.getFeatureName(),
effectInfo.getFeatureType(),
effectInfo.getFeatureBiotype(),
effectInfo.getFeatureChromosome(),
effectInfo.getFeatureStart(),
effectInfo.getFeatureEnd(),
effectInfo.getFeatureStrand(),
effectInfo.getSnpId(),
effectInfo.getAncestral(),
effectInfo.getAlternative(),
effectInfo.getGeneId(),
effectInfo.getTranscriptId(),
effectInfo.getGeneName(),
effectInfo.getConsequenceType(),
effectInfo.getConsequenceTypeObo(),
effectInfo.getConsequenceTypeDesc(),
effectInfo.getConsequenceTypeType(),
effectInfo.getAaPosition(),
effectInfo.getAminoacidChange(),
effectInfo.getCodonChange());
resultsMap.put(new String(r.getRow(), CHARSET_UTF_8), partialResult);
}
}
} catch (InvalidProtocolBufferException e) {
System.err.println(e.getClass().getName() + ": " + e.getMessage());
} catch (IOException e) {
System.err.println(e.getClass().getName() + ": " + e.getMessage());
}
List<Variant> results = new ArrayList<>(resultsMap.values());
return results;
}
/**
* init station code map
*
* @param file station code file path
*/
private static void initStationCode(String file) {
List<String> lstStation = new Vector<>();
strConfig = file + ".process";
File fsP = new File(strConfig);
if (!fsP.exists()) {
try {
fsP.createNewFile();
} catch (IOException e) {
e.printStackTrace();
}
} else {
BufferedReader readp = null;
try {
String line = null;
readp = new BufferedReader(new FileReader(fsP));
while ((line = readp.readLine()) != null) {
processed.add(line);
}
} catch (IOException e) {
e.printStackTrace();
} finally {
if (readp != null) {
try {
readp.close();
} catch (IOException e) {
e.printStackTrace();
}
}
}
}
File fs = new File(file);
BufferedReader reader = null;
try {
String stationName = null;
String stationCode = null;
String line = null;
String[] splitLine = null;
reader = new BufferedReader(new FileReader(fs));
while ((line = reader.readLine()) != null) {
splitLine = line.split("\t");
if (splitLine.length != 2) {
continue;
}
stationName = splitLine[0];
stationCode = splitLine[1];
mapStationCode.put(stationName, stationCode);
lstStation.add(stationName);
}
} catch (FileNotFoundException e) {
e.printStackTrace();
} catch (IOException e) {
e.printStackTrace();
} finally {
if (reader == null) {
try {
reader.close();
} catch (IOException e1) {
e1.printStackTrace();
}
}
}
int len = lstStation.size();
String start = null;
for (int i = 0; i < len; i++) {
start = lstStation.get(i);
for (String end : lstStation) {
if (start.equals(end) || processed.contains(start + ":" + end)) {
continue;
}
lstAllProcessStation.add(new DefaultKeyValue(start, end));
}
}
}
public static List<KeyValue> getAllTrainInfo(Configuration config, String date) {
List<KeyValue> result = new ArrayList<>();
String strJson = null;
BufferedWriter writer = null;
Table table = null;
try (Connection connect = ConnectionFactory.createConnection(config);
Admin admin = connect.getAdmin()) {
TableName tablename = TableName.valueOf(TABLE_NAME);
if (!admin.tableExists(tablename)) {
System.out.println("Table does not exist.");
return null;
}
table = connect.getTable(tablename);
Put put = null;
String start = null;
String end = null;
writer = new BufferedWriter(new FileWriter(new File(strConfig), true));
for (KeyValue item : lstAllProcessStation) {
start = (String) item.getKey();
end = (String) item.getValue();
try {
try {
Thread.sleep(200);
} catch (InterruptedException e1) {
e1.printStackTrace();
}
System.out.println("process : " + start + ":" + end);
strJson = getFromAPIX(mapStationCode.get(start), mapStationCode.get(end), date);
writer.write(start + ":" + end);
writer.newLine();
} catch (Exception e) {
System.out.println(start + ":" + end + "error");
e.printStackTrace();
break;
}
JSONObject jo = new JSONObject(strJson);
if (jo.has("httpstatus") && (jo.getInt("httpstatus") == 200)) {
JSONObject joData = jo.getJSONObject("data");
if (joData.has("flag") && joData.getBoolean("flag")) {
result.add(new DefaultKeyValue(start, end));
// 插入到hbase
String rowkey = start + ":" + end;
put = new Put(rowkey.getBytes());
put.addColumn(
CF_JSON.getBytes(), "json".getBytes(), joData.toString().getBytes("utf-8"));
table.put(put);
System.out.println("start " + start + "\t end " + end + "\t has ticket");
}
}
}
} catch (IOException e) {
e.printStackTrace();
} finally {
if (writer != null) {
try {
writer.flush();
writer.close();
} catch (IOException e) {
e.printStackTrace();
}
}
if (table != null) {
try {
table.close();
} catch (IOException e) {
e.printStackTrace();
}
}
}
return result;
}