private InputStream OpenMultiplePartsWithOffset(FileSystem fs, Path pt, long offset)
throws IOException {
RemoteIterator<LocatedFileStatus> rit = fs.listFiles(pt, false);
Vector<FSDataInputStream> fileHandleList = new Vector<FSDataInputStream>();
while (rit.hasNext()) {
Path path = rit.next().getPath();
String filename =
path.toString().substring(path.getParent().toString().length(), path.toString().length());
if (filename.startsWith("/part-")) {
long filesize = fs.getFileStatus(path).getLen();
if (offset < filesize) {
FSDataInputStream handle = fs.open(path);
if (offset > 0) {
handle.seek(offset);
}
fileHandleList.add(handle);
}
offset -= filesize;
}
}
if (fileHandleList.size() == 1) return fileHandleList.get(0);
else if (fileHandleList.size() > 1) {
Enumeration<FSDataInputStream> enu = fileHandleList.elements();
return new SequenceInputStream(enu);
} else {
System.err.println("Error, no source file loaded. run genSeedDataset.sh fisrt!");
return null;
}
}
public int run(String[] args) throws Exception {
Configuration conf = getConf();
conf.setLong("mapreduce.task.timeout", 10000 * 60 * 60);
Path train_file = new Path(args[0]);
Path test_file = new Path(args[1]);
conf.set("train_file", train_file.getParent().getName());
Job job = new Job(conf, "MapTestID");
job.setJarByClass(MapTestID.class);
job.setOutputKeyClass(Text.class);
job.setOutputValueClass(Text.class);
job.setMapperClass(MapTestID.MapTestIDMap.class);
job.setReducerClass(MapTestID.MapTestIDReduce.class);
job.setNumReduceTasks(300);
FileInputFormat.addInputPath(job, train_file);
FileInputFormat.addInputPath(job, test_file);
FileOutputFormat.setOutputPath(job, new Path(args[2]));
return job.waitForCompletion(true) ? 0 : 1;
}
public void initialize(InputSplit genericSplit, TaskAttemptContext context) throws IOException {
FileSplit split = (FileSplit) genericSplit;
Configuration job = context.getConfiguration();
m_Sb.setLength(0);
m_Start = split.getStart();
m_End = m_Start + split.getLength();
final Path file = split.getPath();
compressionCodecs = new CompressionCodecFactory(job);
final CompressionCodec codec = compressionCodecs.getCodec(file);
// open the file and seek to the m_Start of the split
FileSystem fs = file.getFileSystem(job);
// getFileStatus fileStatus = fs.getFileStatus(split.getPath());
//noinspection deprecation
@SuppressWarnings(value = "deprecated")
long length = fs.getLength(file);
FSDataInputStream fileIn = fs.open(split.getPath());
if (m_Start > 0) fileIn.seek(m_Start);
if (codec != null) {
CompressionInputStream inputStream = codec.createInputStream(fileIn);
m_Input = new BufferedReader(new InputStreamReader(inputStream));
m_End = length;
} else {
m_Input = new BufferedReader(new InputStreamReader(fileIn));
}
m_Current = m_Start;
m_Key = split.getPath().getName();
}
public static String absolutePath(Path p) {
if (p == null) return "";
StringBuilder sb = new StringBuilder();
Path parentPath = p.getParent();
if (parentPath == null) return "/";
sb.append(absolutePath(parentPath));
if (sb.length() > 1) sb.append("/");
sb.append(p.getName());
return sb.toString();
}
/**
* Generate the list of files and make them into FileSplits. This needs to be copied to insert a
* filter on acceptable data
*/
@Override
public List<InputSplit> getSplits(JobContext job) throws IOException {
long minSize = Math.max(getFormatMinSplitSize(), getMinSplitSize(job));
long maxSize = getMaxSplitSize(job);
long desiredMappers =
job.getConfiguration().getLong("org.systemsbiology.jxtandem.DesiredXMLInputMappers", 0);
// generate splits
List<InputSplit> splits = new ArrayList<InputSplit>();
List<FileStatus> fileStatuses = listStatus(job);
boolean forceNumberMappers = fileStatuses.size() == 1;
for (FileStatus file : fileStatuses) {
Path path = file.getPath();
if (!isPathAcceptable(path)) // filter acceptable data
continue;
FileSystem fs = path.getFileSystem(job.getConfiguration());
long length = file.getLen();
BlockLocation[] blkLocations = fs.getFileBlockLocations(file, 0, length);
if ((length != 0) && isSplitable(job, path)) {
long blockSize = file.getBlockSize();
// use desired mappers to force more splits
if (forceNumberMappers && desiredMappers > 0)
maxSize = Math.min(maxSize, (length / desiredMappers));
long splitSize = computeSplitSize(blockSize, minSize, maxSize);
long bytesRemaining = length;
while (withinSlop(splitSize, bytesRemaining)) {
int blkIndex = getBlockIndex(blkLocations, length - bytesRemaining);
splits.add(
new FileSplit(
path, length - bytesRemaining, splitSize, blkLocations[blkIndex].getHosts()));
bytesRemaining -= splitSize;
}
if (bytesRemaining != 0) {
splits.add(
new FileSplit(
path,
length - bytesRemaining,
bytesRemaining,
blkLocations[blkLocations.length - 1].getHosts()));
}
} else if (length != 0) {
splits.add(new FileSplit(path, 0, length, blkLocations[0].getHosts()));
} else {
// Create empty hosts array for zero length files
splits.add(new FileSplit(path, 0, length, new String[0]));
}
}
System.out.println("Total # of splits: " + splits.size());
// LOG.debug("Total # of splits: " + splits.size());
return splits;
}
/**
* Add an file path to the current set of classpath entries It adds the file to cache as well.
*
* @param file Path of the file to be added
* @param conf Configuration that contains the classpath setting
*/
public static void addFileToClassPath(Path file, Configuration conf) throws IOException {
String classpath = conf.get("mapred.job.classpath.files");
conf.set(
"mapred.job.classpath.files",
classpath == null
? file.toString()
: classpath + System.getProperty("path.separator") + file.toString());
URI uri = file.makeQualified(file.getFileSystem(conf)).toUri();
addCacheFile(uri, conf);
}
private static URI addArchiveToClassPathHelper(Path archive, Configuration conf)
throws IOException {
String classpath = conf.get("mapred.job.classpath.archives");
// the scheme/authority use ':' as separator. put the unqualified path in classpath
String archivePath = archive.toUri().getPath();
conf.set(
"mapred.job.classpath.archives",
classpath == null
? archivePath
: classpath + System.getProperty("path.separator") + archivePath);
return archive.makeQualified(archive.getFileSystem(conf)).toUri();
}
@Override
protected boolean isSplitable(JobContext context, Path file) {
String fname = file.getName().toLowerCase();
//noinspection SimplifiableIfStatementf,RedundantIfStatement
if (fname.endsWith(".gz")) return false;
return true;
}
private void setupProcedureStore() throws IOException {
Path testDir = UTIL.getDataTestDir();
FileSystem fs = testDir.getFileSystem(conf);
Path logDir = new Path(testDir, "proc-logs");
System.out.println("Logs directory : " + logDir.toString());
fs.delete(logDir, true);
if ("nosync".equals(syncType)) {
store = new NoSyncWalProcedureStore(conf, fs, logDir);
} else {
store = ProcedureTestingUtility.createWalStore(conf, fs, logDir);
}
store.start(numThreads);
store.recoverLease();
store.load(new ProcedureTestingUtility.LoadCounter());
System.out.println(
"Starting new log : " + store.getActiveLogs().get(store.getActiveLogs().size() - 1));
}
@Override
public void run() {
for (CacheStatus lcacheStatus : toBeDeletedCache) {
synchronized (lcacheStatus) {
Path fullUniqueParentDir =
new Path(lcacheStatus.localizedBaseDir, lcacheStatus.uniqueParentDir);
try {
LOG.info("Deleting local cached path: " + fullUniqueParentDir.toString());
deleteLocalPath(asyncDiskService, fs, fullUniqueParentDir);
// decrement the size of the cache from baseDirSize
deleteCacheInfoUpdate(lcacheStatus);
LOG.info("Removed cache " + lcacheStatus.localizedLoadPath);
} catch (IOException e) {
LOG.warn("Error when deleting " + fullUniqueParentDir, e);
}
}
}
}
protected boolean isPathAcceptable(final Path pPath1) {
String path = pPath1.toString().toLowerCase();
if (path.startsWith("part-r-")) return true;
String extension = getExtension();
if (extension != null && path.endsWith(extension.toLowerCase())) return true;
if (extension != null && path.endsWith(extension.toLowerCase() + ".gz")) return true;
//noinspection SimplifiableIfStatement,RedundantIfStatement
if (extension == null) return true;
return false;
}
// the method which actually copies the caches locally and unjars/unzips them
// and does chmod for the files
private static Path localizeCache(
Configuration conf, URI cache, long confFileStamp, CacheStatus cacheStatus, boolean isArchive)
throws IOException {
FileSystem fs = getFileSystem(cache, conf);
FileSystem localFs = FileSystem.getLocal(conf);
Path parchive = null;
if (isArchive) {
parchive =
new Path(
cacheStatus.localizedLoadPath, new Path(cacheStatus.localizedLoadPath.getName()));
} else {
parchive = cacheStatus.localizedLoadPath;
}
if (!localFs.mkdirs(parchive.getParent())) {
throw new IOException(
"Mkdirs failed to create directory " + cacheStatus.localizedLoadPath.toString());
}
String cacheId = cache.getPath();
fs.copyToLocalFile(new Path(cacheId), parchive);
if (isArchive) {
String tmpArchive = parchive.toString().toLowerCase();
File srcFile = new File(parchive.toString());
File destDir = new File(parchive.getParent().toString());
if (tmpArchive.endsWith(".jar")) {
RunJar.unJar(srcFile, destDir);
} else if (tmpArchive.endsWith(".zip")) {
FileUtil.unZip(srcFile, destDir);
} else if (isTarFile(tmpArchive)) {
FileUtil.unTar(srcFile, destDir);
}
// else will not do anyhting
// and copy the file into the dir as it is
}
long cacheSize = FileUtil.getDU(new File(parchive.getParent().toString()));
cacheStatus.size = cacheSize;
addCacheInfoUpdate(cacheStatus);
// do chmod here
try {
// Setting recursive permission to grant everyone read and execute
Path localDir = new Path(cacheStatus.localizedBaseDir, cacheStatus.uniqueParentDir);
LOG.info("Doing chmod on localdir :" + localDir);
FileUtil.chmod(localDir.toString(), "ugo+rx", true);
} catch (InterruptedException e) {
LOG.warn("Exception in chmod" + e.toString());
}
// update cacheStatus to reflect the newly cached file
cacheStatus.mtime = getTimestamp(conf, cache);
return cacheStatus.localizedLoadPath;
}
/**
* Add an archive path to the current set of classpath entries. It adds the archive to cache as
* well. Intended to be used by user code.
*
* @param archive Path of the archive to be added
* @param conf Configuration that contains the classpath setting
* @param fs FileSystem with respect to which {@code archive} should be interpreted.
*/
public static void addArchiveToClassPath(Path archive, Configuration conf, FileSystem fs)
throws IOException {
String archivepath = archive.toUri().getPath();
String classpath = conf.get("mapred.job.classpath.archives");
conf.set(
"mapred.job.classpath.archives",
classpath == null
? archivepath
: classpath + System.getProperty("path.separator") + archivepath);
URI uri = fs.makeQualified(archive).toUri();
addCacheArchive(uri, conf);
}
public static void recursePath(Configuration conf, Path path, Job job) {
try {
FileSystem fs = path.getFileSystem(conf);
FileStatus[] fstats = fs.listStatus(path);
if (fstats != null) {
for (FileStatus f : fstats) {
Path p = f.getPath();
;
if (fs.isFile(p)) {
// connection times out otherwise
System.err.println("file:" + p.toString());
FileInputFormat.addInputPath(job, p);
} else {
System.err.println("dir:" + p.toString());
recursePath(conf, p, job);
}
}
}
} catch (IOException e) {
// shouldn't be here
throw new RuntimeException(e);
}
}
private static boolean writeChecksum(Path localPath, String checksum) {
try {
java.io.BufferedWriter writer =
Files.newBufferedWriter(
java.nio.file.Paths.get(localPath.toString() + ".crc"), Charset.forName("UTF-8"));
writer.write(checksum);
writer.flush();
writer.close();
return true;
} catch (Exception e) {
logger.warn("Unable to write CRC file!");
return false;
}
}
/**
* open a file for writing
*
* @param hdfsPath !null path -
* @return !null stream
*/
@Override
public OutputStream openFileForWrite(final Path src) {
if (isRunningAsUser()) {
return super.openFileForWrite(src);
}
if (true) throw new UnsupportedOperationException("Fix This"); // ToDo
final FileSystem fs = getDFS();
try {
Path parent = src.getParent();
guaranteeDirectory(parent);
return FileSystem.create(fs, src, FULL_FILE_ACCESS);
} catch (IOException e) {
throw new RuntimeException(e);
}
}
private static boolean checksumLocalTest(Path localPath, String checksum) {
String crcPath = localPath.toString() + ".crc";
if (Files.isReadable(java.nio.file.Paths.get(crcPath))) {
try {
List<String> lines =
Files.readAllLines(java.nio.file.Paths.get(crcPath), Charset.forName("UTF-8"));
for (String line : lines) {
if (line.equals(checksum)) {
return true;
}
}
} catch (Exception e) {
return false;
}
}
return false;
}
private static void createSymlink(
Configuration conf,
URI cache,
CacheStatus cacheStatus,
boolean isArchive,
Path currentWorkDir,
boolean honorSymLinkConf)
throws IOException {
boolean doSymlink = honorSymLinkConf && DistributedCache.getSymlink(conf);
if (cache.getFragment() == null) {
doSymlink = false;
}
String link = currentWorkDir.toString() + Path.SEPARATOR + cache.getFragment();
File flink = new File(link);
if (doSymlink) {
if (!flink.exists()) {
FileUtil.symLink(cacheStatus.localizedLoadPath.toString(), link);
}
}
}
/**
* Delete a local path with asyncDiskService if available, or otherwise synchronously with local
* file system.
*/
private static void deleteLocalPath(
MRAsyncDiskService asyncDiskService, LocalFileSystem fs, Path path) throws IOException {
boolean deleted = false;
if (asyncDiskService != null) {
// Try to delete using asyncDiskService
String localPathToDelete = path.toUri().getPath();
deleted = asyncDiskService.moveAndDeleteAbsolutePath(localPathToDelete);
if (!deleted) {
LOG.warn(
"Cannot find DistributedCache path "
+ localPathToDelete
+ " on any of the asyncDiskService volumes!");
}
}
if (!deleted) {
// If no asyncDiskService, we will delete the files synchronously
fs.delete(path, true);
}
LOG.info("Deleted path " + path);
}
/**
* open a file for reading
*
* @param hdfsPath !null path - probably of an existing file
* @return !null stream
*/
@Override
public InputStream openFileForRead(Path src) {
if (isRunningAsUser()) {
return super.openFileForRead(src);
}
String hdfsPath = src.toString();
if (isFileNameLocal(hdfsPath)) {
try {
return new FileInputStream(hdfsPath); // better be local
} catch (FileNotFoundException e) {
throw new RuntimeException(e);
}
}
if (true) throw new UnsupportedOperationException("Fix This"); // ToDo
final FileSystem fs = getDFS();
try {
return fs.open(src);
} catch (IOException e) {
throw new RuntimeException(e);
}
}
/**
* Get the locally cached file or archive; it could either be previously cached (and valid) or
* copy it from the {@link FileSystem} now.
*
* @param cache the cache to be localized, this should be specified as new
* URI(hdfs://hostname:port/absolute_path_to_file#LINKNAME). If no schema or hostname:port is
* provided the file is assumed to be in the filesystem being used in the Configuration
* @param conf The Confguration file which contains the filesystem
* @param subDir The sub cache Dir where you want to localize the files/archives
* @param fileStatus The file status on the dfs.
* @param isArchive if the cache is an archive or a file. In case it is an archive with a .zip or
* .jar or .tar or .tgz or .tar.gz extension it will be unzipped/unjarred/untarred
* automatically and the directory where the archive is unzipped/unjarred/untarred is returned
* as the Path. In case of a file, the path to the file is returned
* @param confFileStamp this is the hdfs file modification timestamp to verify that the file to be
* cached hasn't changed since the job started
* @param fileLength this is the length of the cache file
* @param currentWorkDir this is the directory where you would want to create symlinks for the
* locally cached files/archives
* @param honorSymLinkConf if this is false, then the symlinks are not created even if conf says
* so (this is required for an optimization in task launches
* @param lDirAllocator LocalDirAllocator of the tracker
* @return the path to directory where the archives are unjarred in case of archives, the path to
* the file where the file is copied locally
* @throws IOException
*/
private static Path getLocalCache(
URI cache,
Configuration conf,
Path subDir,
FileStatus fileStatus,
boolean isArchive,
long confFileStamp,
long fileLength,
Path currentWorkDir,
boolean honorSymLinkConf,
MRAsyncDiskService asyncDiskService,
LocalDirAllocator lDirAllocator)
throws IOException {
String key = getKey(cache, conf, confFileStamp);
CacheStatus lcacheStatus;
Path localizedPath;
synchronized (cachedArchives) {
lcacheStatus = cachedArchives.get(key);
if (lcacheStatus == null) {
// was never localized
Path uniqueParentDir = new Path(subDir, String.valueOf(random.nextLong()));
String cachePath = new Path(uniqueParentDir, makeRelative(cache, conf)).toString();
Path localPath = lDirAllocator.getLocalPathForWrite(cachePath, fileLength, conf);
lcacheStatus =
new CacheStatus(
new Path(localPath.toString().replace(cachePath, "")), localPath, uniqueParentDir);
cachedArchives.put(key, lcacheStatus);
}
lcacheStatus.refcount++;
}
boolean initSuccessful = false;
try {
synchronized (lcacheStatus) {
if (!lcacheStatus.isInited()) {
localizedPath = localizeCache(conf, cache, confFileStamp, lcacheStatus, isArchive);
lcacheStatus.initComplete();
} else {
if (fileStatus != null) {
localizedPath =
checkCacheStatusValidity(
conf, cache, confFileStamp, lcacheStatus, fileStatus, isArchive);
} else {
// if fileStatus is null, then the md5 must be correct
// so there is no need to check for cache validity
localizedPath = lcacheStatus.localizedLoadPath;
}
}
createSymlink(conf, cache, lcacheStatus, isArchive, currentWorkDir, honorSymLinkConf);
}
// try deleting stuff if you can
long size = 0;
int numberSubDir = 0;
synchronized (lcacheStatus) {
synchronized (baseDirSize) {
Long get = baseDirSize.get(lcacheStatus.getBaseDir());
if (get != null) {
size = get.longValue();
} else {
LOG.warn("Cannot find size of baseDir: " + lcacheStatus.getBaseDir());
}
}
synchronized (baseDirNumberSubDir) {
Integer get = baseDirNumberSubDir.get(lcacheStatus.getBaseDir());
if (get != null) {
numberSubDir = get.intValue();
} else {
LOG.warn("Cannot find subdirectories limit of baseDir: " + lcacheStatus.getBaseDir());
}
}
}
// setting the cache size to a default of 10GB
long allowedSize = conf.getLong("local.cache.size", DEFAULT_CACHE_SIZE);
long allowedNumberSubDir =
conf.getLong("local.cache.numbersubdir", DEFAULT_CACHE_SUBDIR_LIMIT);
if (allowedSize < size || allowedNumberSubDir < numberSubDir) {
// try some cache deletions
LOG.debug(
"Start deleting released cache because"
+ " [size, allowedSize, numberSubDir, allowedNumberSubDir] ="
+ " ["
+ size
+ ", "
+ allowedSize
+ ", "
+ numberSubDir
+ ", "
+ allowedNumberSubDir
+ "]");
deleteCache(conf, asyncDiskService);
}
initSuccessful = true;
return localizedPath;
} finally {
if (!initSuccessful) {
synchronized (cachedArchives) {
lcacheStatus.refcount--;
}
}
}
}
/**
* Add a file path to the current set of classpath entries. It adds the file to cache as well.
* Intended to be used by user code.
*
* @deprecated Please use {@link #addFileToClassPath(Path, Configuration, FileSystem)} instead.
* The {@code FileSystem} should be obtained within an appropriate {@code doAs}.
* @param file Path of the file to be added
* @param conf Configuration that contains the classpath setting
*/
@Deprecated
public static void addFileToClassPath(Path file, Configuration conf) throws IOException {
addFileToClassPath(file, conf, file.getFileSystem(conf));
}
public int run(String[] args) throws Exception {
// printUsage();
/*
* SETUP
*/
Configuration argConf = getConf();
Hashtable<String, String> confArg = new Hashtable<String, String>();
setup(confArg, argConf);
Date currentTime = new Date();
Date endDate = new Date(new Long(confArg.get("timestamp_stop")));
Boolean full_run = confArg.get("intermediate").matches("(?i).*true.*");
Boolean quick_add = confArg.get("quick_add").matches("(?i).*true.*");
logger.info("Running GeStore");
// ZooKeeper setup
Configuration config = HBaseConfiguration.create();
zkWatcher = new ZooKeeperWatcher(config, "Testing", new HBaseAdmin(config));
zkInstance =
new ZooKeeper(
ZKConfig.getZKQuorumServersString(config),
config.getInt("zookeeper.session.timeout", -1),
zkWatcher);
if (!confArg.get("task_id").isEmpty()) {
confArg.put("temp_path", confArg.get("temp_path") + confArg.get("task_id"));
}
String lockRequest = confArg.get("file_id");
if (!confArg.get("run_id").isEmpty())
lockRequest = lockRequest + "_" + confArg.get("run_id") + "_";
if (!confArg.get("task_id").isEmpty())
lockRequest = lockRequest + "_" + confArg.get("task_id") + "_";
// Get type of movement
toFrom type_move = checkArgs(confArg);
if (type_move == toFrom.LOCAL2REMOTE && !confArg.get("format").equals("unknown")) {
List<String> arguments = new ArrayList<String>();
arguments.add("-Dinput=" + confArg.get("local_path"));
arguments.add("-Dtable=" + confArg.get("file_id"));
arguments.add("-Dtimestamp=" + confArg.get("timestamp_stop"));
arguments.add("-Dtype=" + confArg.get("format"));
arguments.add("-Dtarget_dir=" + confArg.get("base_path") + "_" + confArg.get("file_id"));
arguments.add("-Dtemp_hdfs_path=" + confArg.get("temp_path"));
arguments.add("-Drun_id=" + confArg.get("run_id"));
if (!confArg.get("run_id").isEmpty()) arguments.add("-Drun_id=" + confArg.get("run_id"));
if (!confArg.get("task_id").isEmpty()) arguments.add("-Dtask_id=" + confArg.get("task_id"));
if (quick_add) arguments.add("-Dquick_add=" + confArg.get("quick_add"));
String lockName = lock(lockRequest);
String[] argumentString = arguments.toArray(new String[arguments.size()]);
adddb.main(argumentString);
unlock(lockName);
System.exit(0);
}
// Database registration
dbutil db_util = new dbutil(config);
db_util.register_database(confArg.get("db_name_files"), true);
db_util.register_database(confArg.get("db_name_runs"), true);
db_util.register_database(confArg.get("db_name_updates"), true);
FileSystem hdfs = FileSystem.get(config);
FileSystem localFS = FileSystem.getLocal(config);
// Get source type
confArg.put("source", getSource(db_util, confArg.get("db_name_files"), confArg.get("file_id")));
confArg.put(
"database", isDatabase(db_util, confArg.get("db_name_files"), confArg.get("file_id")));
if (!confArg.get("source").equals("local")
&& type_move == toFrom.REMOTE2LOCAL
&& !confArg.get("timestamp_stop").equals(Integer.toString(Integer.MAX_VALUE))) {
confArg.put("timestamp_stop", Long.toString(latestVersion(confArg, db_util)));
}
/*
* Get previous timestamp
*/
Get run_id_get = new Get(confArg.get("run_id").getBytes());
Result run_get = db_util.doGet(confArg.get("db_name_runs"), run_id_get);
KeyValue run_file_prev =
run_get.getColumnLatest(
"d".getBytes(), (confArg.get("file_id") + "_db_timestamp").getBytes());
String last_timestamp = new String("0");
if (null != run_file_prev && !confArg.get("source").equals("local")) {
long last_timestamp_real = run_file_prev.getTimestamp();
Long current_timestamp = new Long(confArg.get("timestamp_real"));
if ((current_timestamp - last_timestamp_real) > 36000) {
last_timestamp = new String(run_file_prev.getValue());
Integer lastTimestamp = new Integer(last_timestamp);
lastTimestamp += 1;
last_timestamp = lastTimestamp.toString();
logger.info("Last timestamp: " + last_timestamp + " End data: " + endDate);
Date last_run = new Date(run_file_prev.getTimestamp());
if (last_run.before(endDate) && !full_run) {
confArg.put("timestamp_start", last_timestamp);
}
}
}
Integer tse = new Integer(confArg.get("timestamp_stop"));
Integer tss = new Integer(confArg.get("timestamp_start"));
if (tss > tse) {
logger.info("No new version of requested file.");
return 0;
}
/*
* Generate file
*/
String lockName = lock(lockRequest);
Get file_id_get = new Get(confArg.get("file_id").getBytes());
Result file_get = db_util.doGet(confArg.get("db_name_files"), file_id_get);
if (!file_get.isEmpty()) {
boolean found =
hasFile(
db_util,
hdfs,
confArg.get("db_name_files"),
confArg.get("file_id"),
getFullPath(confArg));
if (confArg.get("source").equals("fullfile")) {
found = false;
}
String filenames_put =
getFileNames(
db_util, confArg.get("db_name_files"), confArg.get("file_id"), getFullPath(confArg));
// Filename not found in file database
if (!found && type_move == toFrom.REMOTE2LOCAL) {
if (!confArg.get("source").equals("local")) {
// Generate intermediate file
if (getFile(hdfs, confArg, db_util) == null) {
unlock(lockName);
return 1;
}
// Put generated file into file database
if (!confArg.get("format").equals("fullfile")) {
putFileEntry(
db_util,
hdfs,
confArg.get("db_name_files"),
confArg.get("file_id"),
confArg.get("full_file_name"),
confArg.get("source"));
}
} else {
logger.warn("Remote file not found, and cannot be generated! File: " + confArg);
unlock(lockName);
return 1;
}
}
} else {
if (type_move == toFrom.REMOTE2LOCAL) {
logger.warn("Remote file not found, and cannot be generated.");
unlock(lockName);
return 1;
}
}
/*
* Copy file
* Update tables
*/
if (type_move == toFrom.LOCAL2REMOTE) {
if (!confArg.get("format").equals("fullfile")) {
putFileEntry(
db_util,
hdfs,
confArg.get("db_name_files"),
confArg.get("file_id"),
getFullPath(confArg),
confArg.get("source"));
}
putRunEntry(
db_util,
confArg.get("db_name_runs"),
confArg.get("run_id"),
confArg.get("file_id"),
confArg.get("type"),
confArg.get("timestamp_real"),
confArg.get("timestamp_stop"),
getFullPath(confArg),
confArg.get("delimiter"));
hdfs.copyFromLocalFile(new Path(confArg.get("local_path")), new Path(getFullPath(confArg)));
} else if (type_move == toFrom.REMOTE2LOCAL) {
FileStatus[] files = hdfs.globStatus(new Path(getFullPath(confArg) + "*"));
putRunEntry(
db_util,
confArg.get("db_name_runs"),
confArg.get("run_id"),
confArg.get("file_id"),
confArg.get("type"),
confArg.get("timestamp_real"),
confArg.get("timestamp_stop"),
getFullPath(confArg),
confArg.get("delimiter"));
unlock(lockName);
for (FileStatus file : files) {
Path cur_file = file.getPath();
Path cur_local_path =
new Path(new String(confArg.get("local_path") + confArg.get("file_id")));
String suffix = getSuffix(getFileName(confArg), cur_file.getName());
if (suffix.length() > 0) {
cur_local_path = cur_local_path.suffix(new String("." + suffix));
}
if (confArg.get("copy").equals("true")) {
String crc = hdfs.getFileChecksum(cur_file).toString();
if (checksumLocalTest(cur_local_path, crc)) {
continue;
} else {
hdfs.copyToLocalFile(cur_file, cur_local_path);
writeChecksum(cur_local_path, crc);
}
} else {
System.out.println(cur_local_path + "\t" + cur_file);
}
}
}
unlock(lockName);
return 0;
}
// Information needed to get a single file:
// BASE_PATH, FILE_ID, TIMESTAMP_START, TIMESTAMP_STOP, SOURCE, FILESYSTEM
private static Vector<Path> getFile(
FileSystem fs, Hashtable<String, String> config, dbutil db_util) throws Exception {
Long latestVersion = latestVersion(config, db_util);
try {
config.put("timestamp_start", config.get("timestamp_start"));
config.put("timestamp_real", latestVersion.toString());
config.put("timestamp_stop", latestVersion.toString());
} catch (Exception E) {
logger.error("Tryign to get file that is impossible to generate: " + getFullPath(config));
return null;
}
if (Integer.parseInt(config.get("timestamp_start"))
> Integer.parseInt(config.get("timestamp_stop"))) {
return null;
}
logger.debug(
"Getting DB for timestamp "
+ config.get("timestamp_start")
+ " to "
+ config.get("timestamp_stop"));
String final_result = getFullPath(config);
String temp_path_base =
config.get("local_temp_path")
+ "_"
+ config.get("task_id")
+ "_"
+ config.get("run_id")
+ "/";
Path newPath = new Path(final_result + "*");
Vector<Path> ret_path = new Vector<Path>();
String lockName = lock(final_result.replaceAll("/", "_"));
if (fs.globStatus(newPath).length != 0) {
ret_path.add(newPath);
unlock(lockName);
config.put("full_file_name", final_result);
return ret_path;
} else {
if (!config.get("source").equals("local")) {
config.put("temp_path_base", temp_path_base);
config.put("timestamp_start", config.get("timestamp_start"));
config.put("timestamp_real", latestVersion.toString());
config.put("timestamp_stop", latestVersion.toString());
Class<?> sourceClass =
Class.forName("org.gestore.plugin.source." + config.get("source") + "Source");
Method process_data = sourceClass.getMethod("process", Hashtable.class, FileSystem.class);
Object processor = sourceClass.newInstance();
Object retVal;
try {
retVal = process_data.invoke(processor, config, fs);
} catch (InvocationTargetException E) {
Throwable exception = E.getTargetException();
logger.error("Unable to call method in child class: " + exception.toString());
exception.printStackTrace(System.out);
unlock(lockName);
return null;
}
FileStatus[] files = (FileStatus[]) retVal;
if (files == null) {
logger.error("Error getting files, no files returned");
return null;
}
for (FileStatus file : files) {
Path cur_file = file.getPath();
Path cur_local_path = new Path(temp_path_base + config.get("file_id"));
String suffix = getSuffix(config.get("file_id"), cur_file.getName());
cur_local_path = cur_local_path.suffix(suffix);
Path res_path = new Path(new String(final_result + suffix));
logger.debug("Moving file" + cur_file.toString() + " to " + res_path.toString());
if (config.get("copy").equals("true")) {
fs.moveFromLocalFile(cur_file, res_path);
} else {
fs.rename(cur_file, res_path);
}
}
config.put("full_file_name", final_result);
}
}
unlock(lockName);
return ret_path;
}
/**
* Add an archive path to the current set of classpath entries. It adds the archive to cache as
* well. Intended to be used by user code.
*
* @deprecated Please use {@link #addArchiveToClassPath(Path, Configuration, FileSystem)} instead.
* The {@code FileSystem} should be obtained within an appropriate {@code doAs}.
* @param archive Path of the archive to be added
* @param conf Configuration that contains the classpath setting
*/
@Deprecated
public static void addArchiveToClassPath(Path archive, Configuration conf) throws IOException {
addArchiveToClassPath(archive, conf, archive.getFileSystem(conf));
}