/** Copy FileSystem files to local files. */
public static boolean copy(
FileSystem srcFS, Path src, File dst, boolean deleteSource, Configuration conf)
throws IOException {
if (srcFS.getFileStatus(src).isDir()) {
if (!dst.mkdirs()) {
return false;
}
FileStatus contents[] = srcFS.listStatus(src);
for (int i = 0; i < contents.length; i++) {
copy(
srcFS,
contents[i].getPath(),
new File(dst, contents[i].getPath().getName()),
deleteSource,
conf);
}
} else if (srcFS.isFile(src)) {
InputStream in = srcFS.open(src);
IOUtils.copyBytes(in, new FileOutputStream(dst), conf);
} else {
throw new IOException(src.toString() + ": No such file or directory");
}
if (deleteSource) {
return srcFS.delete(src, true);
} else {
return true;
}
}
public void testFormat() throws Exception {
localFs = FileSystem.getLocal(defaultConf);
localFs.delete(workDir, true);
Job job = new Job(new Configuration(defaultConf));
Path file = new Path(workDir, "test.txt");
int seed = new Random().nextInt();
Random random = new Random(seed);
// for a variety of lengths
for (int length = 0; length < MAX_LENGTH; length += random.nextInt(MAX_LENGTH / 10) + 1) {
// create a file with length entries
Writer writer = new OutputStreamWriter(localFs.create(file));
try {
MyClass mc = new MyClass();
for (int i = 0; i < length; i++) {
mc.s = Integer.toString(i);
mc.v = i;
byte[] raw = MessagePack.pack(mc);
byte[] b64e = base64_.encodeBase64(raw);
byte[] b64d = base64_.decode(b64e);
MyClass mc2 = MessagePack.unpack(b64d, mc.getClass());
assertEquals(mc.s, mc2.s);
assertEquals(mc.v, mc2.v);
writer.write(base64_.encodeToString(raw));
}
} finally {
writer.close();
}
checkFormat(job);
}
}
public static void main(String[] args) throws Exception {
final String NAME_NODE = "hdfs://sandbox.hortonworks.com:8020";
Configuration conf = new Configuration();
Job job = Job.getInstance(conf);
job.setJarByClass(WordCount.class);
job.setMapOutputKeyClass(Text.class);
job.setMapOutputValueClass(IntWritable.class);
job.setOutputKeyClass(IntWritable.class);
job.setOutputValueClass(NullWritable.class);
if (args.length > 2) {
job.setNumReduceTasks(Integer.parseInt(args[2]));
}
job.setMapperClass(CountMapper.class);
job.setReducerClass(CountReducer.class);
job.setJarByClass(WordCount.class);
job.setNumReduceTasks(1);
FileInputFormat.addInputPath(job, new Path(args[0] + "data/plot_summaries.txt"));
FileSystem fs = FileSystem.get(conf);
// handle (e.g. delete) existing output path
Path outputDestination = new Path(args[0] + args[1]);
if (fs.exists(outputDestination)) {
fs.delete(outputDestination, true);
}
// set output path & start job1
FileOutputFormat.setOutputPath(job, outputDestination);
int jobCompletionStatus = job.waitForCompletion(true) ? 0 : 1;
}
/** Copy local files to a FileSystem. */
public static boolean copy(
File src, FileSystem dstFS, Path dst, boolean deleteSource, Configuration conf)
throws IOException {
dst = checkDest(src.getName(), dstFS, dst, false);
if (src.isDirectory()) {
if (!dstFS.mkdirs(dst)) {
return false;
}
File contents[] = listFiles(src);
for (int i = 0; i < contents.length; i++) {
copy(contents[i], dstFS, new Path(dst, contents[i].getName()), deleteSource, conf);
}
} else if (src.isFile()) {
InputStream in = null;
OutputStream out = null;
try {
in = new FileInputStream(src);
out = dstFS.create(dst);
IOUtils.copyBytes(in, out, conf);
} catch (IOException e) {
IOUtils.closeStream(out);
IOUtils.closeStream(in);
throw e;
}
} else {
throw new IOException(src.toString() + ": No such file or directory");
}
if (deleteSource) {
return FileUtil.fullyDelete(src);
} else {
return true;
}
}
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 void inject(Path crawlDb, Path urlDir) throws IOException {
SimpleDateFormat sdf = new SimpleDateFormat("yyyy-MM-dd HH:mm:ss");
long start = System.currentTimeMillis();
if (LOG.isInfoEnabled()) {
LOG.info("Injector: starting at " + sdf.format(start));
LOG.info("Injector: crawlDb: " + crawlDb);
LOG.info("Injector: urlDir: " + urlDir);
}
Path tempDir =
new Path(
getConf().get("mapred.temp.dir", ".")
+ "/inject-temp-"
+ Integer.toString(new Random().nextInt(Integer.MAX_VALUE)));
// map text input file to a <url,CrawlDatum> file
if (LOG.isInfoEnabled()) {
LOG.info("Injector: Converting injected urls to crawl db entries.");
}
JobConf sortJob = new NutchJob(getConf());
sortJob.setJobName("inject " + urlDir);
FileInputFormat.addInputPath(sortJob, urlDir);
sortJob.setMapperClass(InjectMapper.class);
FileOutputFormat.setOutputPath(sortJob, tempDir);
sortJob.setOutputFormat(SequenceFileOutputFormat.class);
sortJob.setOutputKeyClass(Text.class);
sortJob.setOutputValueClass(CrawlDatum.class);
sortJob.setLong("injector.current.time", System.currentTimeMillis());
RunningJob mapJob = JobClient.runJob(sortJob);
long urlsInjected = mapJob.getCounters().findCounter("injector", "urls_injected").getValue();
long urlsFiltered = mapJob.getCounters().findCounter("injector", "urls_filtered").getValue();
LOG.info("Injector: total number of urls rejected by filters: " + urlsFiltered);
LOG.info(
"Injector: total number of urls injected after normalization and filtering: "
+ urlsInjected);
// merge with existing crawl db
if (LOG.isInfoEnabled()) {
LOG.info("Injector: Merging injected urls into crawl db.");
}
JobConf mergeJob = CrawlDb.createJob(getConf(), crawlDb);
FileInputFormat.addInputPath(mergeJob, tempDir);
mergeJob.setReducerClass(InjectReducer.class);
JobClient.runJob(mergeJob);
CrawlDb.install(mergeJob, crawlDb);
// clean up
FileSystem fs = FileSystem.get(getConf());
fs.delete(tempDir, true);
long end = System.currentTimeMillis();
LOG.info(
"Injector: finished at "
+ sdf.format(end)
+ ", elapsed: "
+ TimingUtil.elapsedTime(start, end));
}
// 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;
}
protected void setup(Context context) throws IOException, InterruptedException {
super.setup(context);
Path pt = new Path("/user/yao/query/query");
FileSystem fs = FileSystem.get(new Configuration());
BufferedReader br = new BufferedReader(new InputStreamReader(fs.open(pt)));
String line = br.readLine();
String[] keywords = line.split(",");
k0 = keywords[0];
k1 = keywords[1];
k2 = keywords[2];
br.close();
}
/**
* Opens an FSDataInputStream at the indicated Path.
*
* @param f the file name to open
* @param bufferSize the size of the buffer to be used.
*/
@Override
public FSDataInputStream open(Path f, int bufferSize) throws IOException {
FileSystem fs;
InputStream in;
if (verifyChecksum) {
fs = this;
in = new ChecksumFSInputChecker(this, f, bufferSize);
} else {
fs = getRawFileSystem();
in = fs.open(f, bufferSize);
}
return new FSDataBoundedInputStream(fs, f, in);
}
private static Path checkDest(String srcName, FileSystem dstFS, Path dst, boolean overwrite)
throws IOException {
if (dstFS.exists(dst)) {
FileStatus sdst = dstFS.getFileStatus(dst);
if (sdst.isDir()) {
if (null == srcName) {
throw new IOException("Target " + dst + " is a directory");
}
return checkDest(null, dstFS, new Path(dst, srcName), overwrite);
} else if (!overwrite) {
throw new IOException("Target " + dst + " already exists");
}
}
return dst;
}
public void configure(JobConf conf) {
numberOfCenters = Integer.valueOf(conf.get("numberOfCenters"));
centersDirectory = conf.get("centersReadDirectory");
try {
Configuration c = new Configuration();
FileSystem fs = FileSystem.get(c);
for (int index = 0; index < numberOfCenters; ++index) {
SequenceFile.Reader reader =
new SequenceFile.Reader(fs, new Path(centersDirectory + "/centers/" + index), c);
LongWritable key = new LongWritable();
Point value = new Point();
reader.next(key, value);
Point center = (Point) value;
centers.add(center);
reader.close();
}
} catch (IOException e) {
// do nothing
// I hope this doesn't happen
System.out.println("well, damn.");
e.printStackTrace();
}
}
public void testAbort() throws IOException {
JobConf job = new JobConf();
setConfForFileOutputCommitter(job);
JobContext jContext = new JobContextImpl(job, taskID.getJobID());
TaskAttemptContext tContext = new TaskAttemptContextImpl(job, taskID);
FileOutputCommitter committer = new FileOutputCommitter();
FileOutputFormat.setWorkOutputPath(job, committer.getTempTaskOutputPath(tContext));
// do setup
committer.setupJob(jContext);
committer.setupTask(tContext);
String file = "test.txt";
// A reporter that does nothing
Reporter reporter = Reporter.NULL;
// write output
FileSystem localFs = FileSystem.getLocal(job);
TextOutputFormat theOutputFormat = new TextOutputFormat();
RecordWriter theRecordWriter = theOutputFormat.getRecordWriter(localFs, job, file, reporter);
writeOutput(theRecordWriter, reporter);
// do abort
committer.abortTask(tContext);
File expectedFile =
new File(new Path(committer.getTempTaskOutputPath(tContext), file).toString());
assertFalse("task temp dir still exists", expectedFile.exists());
committer.abortJob(jContext, JobStatus.State.FAILED);
expectedFile = new File(new Path(outDir, FileOutputCommitter.TEMP_DIR_NAME).toString());
assertFalse("job temp dir still exists", expectedFile.exists());
assertEquals("Output directory not empty", 0, new File(outDir.toString()).listFiles().length);
FileUtil.fullyDelete(new File(outDir.toString()));
}
private void insertFile(
Path apkPath, Map<String, String> zip_properties, File insert, String method, Path location)
throws AndrolibException, IOException {
// ZipFileSystem only writes at .close()
// http://mail.openjdk.java.net/pipermail/nio-dev/2012-July/001764.html
try (FileSystem fs = FileSystems.newFileSystem(apkPath, null)) {
Path root = fs.getPath("/");
// in order to get the path relative to the zip, we strip off the absolute path, minus what we
// already have in the zip. thus /var/files/apktool/apk/unknown/folder/file => /folder/file
Path dest =
fs.getPath(root.toString(), insert.getAbsolutePath().replace(location.toString(), ""));
Path newFile = Paths.get(insert.getAbsolutePath());
Files.copy(newFile, dest, StandardCopyOption.REPLACE_EXISTING);
fs.close();
}
}
public static void main(String[] args) throws Exception {
try {
String localSrc = args[0];
String dst = args[1];
InputStream in = new BufferedInputStream(new FileInputStream(localSrc));
Configuration conf = new Configuration();
FileSystem fs = FileSystem.get(URI.create(dst), conf);
OutputStream out = fs.create(new Path(dst));
IOUtils.copyBytes(in, out, 4096, true);
System.out.print("success");
} catch (Exception e) {
System.out.print("fail" + e.toString());
}
}
/** Start the JobTracker process, listen on the indicated port */
JobTracker(Configuration conf) throws IOException {
//
// Grab some static constants
//
maxCurrentTasks = conf.getInt("mapred.tasktracker.tasks.maximum", 2);
RETIRE_JOB_INTERVAL = conf.getLong("mapred.jobtracker.retirejob.interval", 24 * 60 * 60 * 1000);
RETIRE_JOB_CHECK_INTERVAL = conf.getLong("mapred.jobtracker.retirejob.check", 60 * 1000);
TASK_ALLOC_EPSILON = conf.getFloat("mapred.jobtracker.taskalloc.loadbalance.epsilon", 0.2f);
PAD_FRACTION = conf.getFloat("mapred.jobtracker.taskalloc.capacitypad", 0.1f);
MIN_SLOTS_FOR_PADDING = 3 * maxCurrentTasks;
// This is a directory of temporary submission files. We delete it
// on startup, and can delete any files that we're done with
this.conf = conf;
JobConf jobConf = new JobConf(conf);
this.systemDir = jobConf.getSystemDir();
this.fs = FileSystem.get(conf);
FileUtil.fullyDelete(fs, systemDir);
fs.mkdirs(systemDir);
// Same with 'localDir' except it's always on the local disk.
jobConf.deleteLocalFiles(SUBDIR);
// Set ports, start RPC servers, etc.
InetSocketAddress addr = getAddress(conf);
this.localMachine = addr.getHostName();
this.port = addr.getPort();
this.interTrackerServer = RPC.getServer(this, addr.getPort(), 10, false, conf);
this.interTrackerServer.start();
Properties p = System.getProperties();
for (Iterator it = p.keySet().iterator(); it.hasNext(); ) {
String key = (String) it.next();
String val = (String) p.getProperty(key);
LOG.info("Property '" + key + "' is " + val);
}
this.infoPort = conf.getInt("mapred.job.tracker.info.port", 50030);
this.infoServer = new JobTrackerInfoServer(this, infoPort);
this.infoServer.start();
this.startTime = System.currentTimeMillis();
new Thread(this.expireTrackers).start();
new Thread(this.retireJobs).start();
new Thread(this.initJobs).start();
}
private void insertFolder(
Path apkPath, Map<String, String> zip_properties, File insert, String method, Path location)
throws AndrolibException, IOException {
try (FileSystem fs = FileSystems.newFileSystem(apkPath, null)) {
Path root = fs.getPath("/");
Path dest =
fs.getPath(root.toString(), insert.getAbsolutePath().replace(location.toString(), ""));
Path parent = dest.normalize();
// check for folder existing in apkFileSystem
if (parent != null && Files.notExists(parent)) {
if (!Files.isDirectory(parent, LinkOption.NOFOLLOW_LINKS)) {
Files.createDirectories(parent);
}
}
fs.close();
}
}
public static boolean copy(
FileSystem srcFS,
Path[] srcs,
FileSystem dstFS,
Path dst,
boolean deleteSource,
boolean overwrite,
Configuration conf)
throws IOException {
boolean gotException = false;
boolean returnVal = true;
StringBuffer exceptions = new StringBuffer();
if (srcs.length == 1) return copy(srcFS, srcs[0], dstFS, dst, deleteSource, overwrite, conf);
// Check if dest is directory
if (!dstFS.exists(dst)) {
throw new IOException("`" + dst + "': specified destination directory " + "doest not exist");
} else {
FileStatus sdst = dstFS.getFileStatus(dst);
if (!sdst.isDir())
throw new IOException(
"copying multiple files, but last argument `" + dst + "' is not a directory");
}
for (Path src : srcs) {
try {
if (!copy(srcFS, src, dstFS, dst, deleteSource, overwrite, conf)) returnVal = false;
} catch (IOException e) {
gotException = true;
exceptions.append(e.getMessage());
exceptions.append("\n");
}
}
if (gotException) {
throw new IOException(exceptions.toString());
}
return returnVal;
}
/**
* Clear the entire contents of the cache and delete the backing files. This should only be used
* when the server is reinitializing, because the users are going to lose their files.
*/
public static void purgeCache(Configuration conf, MRAsyncDiskService service) throws IOException {
synchronized (cachedArchives) {
LocalFileSystem localFs = FileSystem.getLocal(conf);
for (Map.Entry<String, CacheStatus> f : cachedArchives.entrySet()) {
try {
deleteLocalPath(service, localFs, f.getValue().localizedLoadPath);
} catch (IOException ie) {
LOG.debug("Error cleaning up cache", ie);
}
}
cachedArchives.clear();
}
}
/** Copy all files in a directory to one output file (merge). */
public static boolean copyMerge(
FileSystem srcFS,
Path srcDir,
FileSystem dstFS,
Path dstFile,
boolean deleteSource,
Configuration conf,
String addString)
throws IOException {
dstFile = checkDest(srcDir.getName(), dstFS, dstFile, false);
if (!srcFS.getFileStatus(srcDir).isDir()) return false;
OutputStream out = dstFS.create(dstFile);
try {
FileStatus contents[] = srcFS.listStatus(srcDir);
for (int i = 0; i < contents.length; i++) {
if (!contents[i].isDir()) {
InputStream in = srcFS.open(contents[i].getPath());
try {
IOUtils.copyBytes(in, out, conf, false);
if (addString != null) out.write(addString.getBytes("UTF-8"));
} finally {
in.close();
}
}
}
} finally {
out.close();
}
if (deleteSource) {
return srcFS.delete(srcDir, true);
} else {
return true;
}
}
/**
* The src file is under FS, and the dst is on the local disk. Copy it from FS control to the
* local dst name. If src and dst are directories, the copyCrc parameter determines whether to
* copy CRC files.
*/
public void copyToLocalFile(Path src, Path dst, boolean copyCrc) throws IOException {
if (!fs.isDirectory(src)) { // source is a file
fs.copyToLocalFile(src, dst);
FileSystem localFs = getLocal(getConf()).getRawFileSystem();
if (localFs.isDirectory(dst)) {
dst = new Path(dst, src.getName());
}
dst = getChecksumFile(dst);
if (localFs.exists(dst)) { // remove old local checksum file
localFs.delete(dst, true);
}
Path checksumFile = getChecksumFile(src);
if (copyCrc && fs.exists(checksumFile)) { // copy checksum file
fs.copyToLocalFile(checksumFile, dst);
}
} else {
FileStatus[] srcs = listStatus(src);
for (FileStatus srcFile : srcs) {
copyToLocalFile(srcFile.getPath(), new Path(dst, srcFile.getPath().getName()), copyCrc);
}
}
}
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);
}
}
public static void extractQueryFeatures2HDFS(String filename, Job job) throws IOException {
// Read the local image.jpg as a Mat
Mat query_mat_float =
Highgui.imread(LOCAL_USER_DIR + ID + INPUT + "/" + filename, CvType.CV_32FC3);
// Convert RGB to GRAY
Mat query_gray = new Mat();
Imgproc.cvtColor(query_mat_float, query_gray, Imgproc.COLOR_RGB2GRAY);
// Convert the float type to unsigned integer(required by SIFT)
Mat query_mat_byte = new Mat();
query_gray.convertTo(query_mat_byte, CvType.CV_8UC3);
// // Resize the image to 1/FACTOR both width and height
// Mat query_mat_byte = FeatureExtraction.resize(query_mat_byte);
// Extract the feature from the (Mat)image
Mat query_features = FeatureExtraction.extractFeature(query_mat_byte);
System.out.println(PREFIX + "Extracting the query image feature...");
System.out.println("query_mat(float,color):" + query_mat_float);
System.out.println("query_mat(float,gray):" + query_gray);
System.out.println("query_mat(byte,gray):" + query_mat_byte);
System.out.println("query_mat_features:" + query_features);
System.out.println();
// Store the feature to the hdfs in order to use it later in different map tasks
System.out.println(PREFIX + "Generating the feature file for the query image in HDFS...");
FileSystem fs = FileSystem.get(job.getConfiguration());
String featureFileName = filename.substring(0, filename.lastIndexOf(".")) + ".json";
FSDataOutputStream fsDataOutputStream =
fs.create(new Path(HDFS_HOME + USER + ID + INPUT + "/" + featureFileName));
BufferedWriter bw =
new BufferedWriter(new OutputStreamWriter(fsDataOutputStream, StandardCharsets.UTF_8));
bw.write(FeatureExtraction.mat2json(query_features));
bw.close();
System.out.println(PREFIX + "Query feature extraction finished...");
System.out.println();
}
public BufferedReader loadDataFromFile(String filepath, long offset) {
try {
Path pt = new Path(filepath);
FileSystem fs = FileSystem.get(fsConf);
InputStreamReader isr;
if (fs.isDirectory(pt)) { // multiple parts
isr = new InputStreamReader(OpenMultiplePartsWithOffset(fs, pt, offset));
} else { // single file
FSDataInputStream fileHandler = fs.open(pt);
if (offset > 0) fileHandler.seek(offset);
isr = new InputStreamReader(fileHandler);
}
BufferedReader reader = new BufferedReader(isr);
if (offset > 0) reader.readLine(); // skip first line in case of seek
return reader;
} catch (FileNotFoundException e) {
e.printStackTrace();
} catch (IOException e) {
e.printStackTrace();
}
assert false : "Should not reach here!";
return null;
}
/*
* Returns the relative path of the dir this cache will be localized in
* relative path that this cache will be localized in. For
* hdfs://hostname:port/absolute_path -- the relative path is
* hostname/absolute path -- if it is just /absolute_path -- then the
* relative path is hostname of DFS this mapred cluster is running
* on/absolute_path
*/
public static String makeRelative(URI cache, Configuration conf) throws IOException {
String host = cache.getHost();
if (host == null) {
host = cache.getScheme();
}
if (host == null) {
URI defaultUri = FileSystem.get(conf).getUri();
host = defaultUri.getHost();
if (host == null) {
host = defaultUri.getScheme();
}
}
String path = host + cache.getPath();
path = path.replace(":/", "/"); // remove windows device colon
return path;
}
/** Copy files between FileSystems. */
public static boolean copy(
FileSystem srcFS,
Path src,
FileSystem dstFS,
Path dst,
boolean deleteSource,
boolean overwrite,
Configuration conf)
throws IOException {
dst = checkDest(src.getName(), dstFS, dst, overwrite);
if (srcFS.getFileStatus(src).isDir()) {
checkDependencies(srcFS, src, dstFS, dst);
if (!dstFS.mkdirs(dst)) {
return false;
}
FileStatus contents[] = srcFS.listStatus(src);
for (int i = 0; i < contents.length; i++) {
copy(
srcFS,
contents[i].getPath(),
dstFS,
new Path(dst, contents[i].getPath().getName()),
deleteSource,
overwrite,
conf);
}
} else if (srcFS.isFile(src)) {
InputStream in = null;
OutputStream out = null;
try {
in = srcFS.open(src);
out = dstFS.create(dst, overwrite);
IOUtils.copyBytes(in, out, conf, true);
} catch (IOException e) {
IOUtils.closeStream(out);
IOUtils.closeStream(in);
throw e;
}
} else {
throw new IOException(src.toString() + ": No such file or directory");
}
if (deleteSource) {
return srcFS.delete(src, true);
} else {
return true;
}
}
private static Path checkCacheStatusValidity(
Configuration conf,
URI cache,
long confFileStamp,
CacheStatus cacheStatus,
FileStatus fileStatus,
boolean isArchive)
throws IOException {
FileSystem fs = FileSystem.get(cache, conf);
// Has to be
if (!ifExistsAndFresh(conf, fs, cache, confFileStamp, cacheStatus, fileStatus)) {
throw new IOException(
"Stale cache file: " + cacheStatus.localizedLoadPath + " for cache-file: " + cache);
}
LOG.info(
String.format(
"Using existing cache of %s->%s", cache.toString(), cacheStatus.localizedLoadPath));
return cacheStatus.localizedLoadPath;
}
private static void deleteCache(Configuration conf, MRAsyncDiskService asyncDiskService)
throws IOException {
List<CacheStatus> deleteSet = new LinkedList<CacheStatus>();
// try deleting cache Status with refcount of zero
synchronized (cachedArchives) {
for (Iterator<String> it = cachedArchives.keySet().iterator(); it.hasNext(); ) {
String cacheId = (String) it.next();
CacheStatus lcacheStatus = cachedArchives.get(cacheId);
if (lcacheStatus.refcount == 0) {
// delete this cache entry from the global list
// and mark the localized file for deletion
deleteSet.add(lcacheStatus);
it.remove();
}
}
}
// do the deletion asynchronously, after releasing the global lock
Thread cacheFileCleaner =
new Thread(new CacheFileCleanTask(asyncDiskService, FileSystem.getLocal(conf), deleteSet));
cacheFileCleaner.start();
}
@SuppressWarnings("unchecked")
public void testCommitter() throws Exception {
JobConf job = new JobConf();
setConfForFileOutputCommitter(job);
JobContext jContext = new JobContextImpl(job, taskID.getJobID());
TaskAttemptContext tContext = new TaskAttemptContextImpl(job, taskID);
FileOutputCommitter committer = new FileOutputCommitter();
FileOutputFormat.setWorkOutputPath(job, committer.getTempTaskOutputPath(tContext));
committer.setupJob(jContext);
committer.setupTask(tContext);
String file = "test.txt";
// A reporter that does nothing
Reporter reporter = Reporter.NULL;
// write output
FileSystem localFs = FileSystem.getLocal(job);
TextOutputFormat theOutputFormat = new TextOutputFormat();
RecordWriter theRecordWriter = theOutputFormat.getRecordWriter(localFs, job, file, reporter);
writeOutput(theRecordWriter, reporter);
// do commit
committer.commitTask(tContext);
committer.commitJob(jContext);
// validate output
File expectedFile = new File(new Path(outDir, file).toString());
StringBuffer expectedOutput = new StringBuffer();
expectedOutput.append(key1).append('\t').append(val1).append("\n");
expectedOutput.append(val1).append("\n");
expectedOutput.append(val2).append("\n");
expectedOutput.append(key2).append("\n");
expectedOutput.append(key1).append("\n");
expectedOutput.append(key2).append('\t').append(val2).append("\n");
String output = UtilsForTests.slurp(expectedFile);
assertEquals(output, expectedOutput.toString());
FileUtil.fullyDelete(new File(outDir.toString()));
}
public void testFormat() throws Exception {
JobConf job = new JobConf(conf);
FileSystem fs = FileSystem.getLocal(conf);
Path dir = new Path(System.getProperty("test.build.data", ".") + "/mapred");
Path file = new Path(dir, "test.seq");
Reporter reporter = Reporter.NULL;
int seed = new Random().nextInt();
// LOG.info("seed = "+seed);
Random random = new Random(seed);
fs.delete(dir, true);
FileInputFormat.setInputPaths(job, dir);
// for a variety of lengths
for (int length = 0; length < MAX_LENGTH; length += random.nextInt(MAX_LENGTH / 10) + 1) {
// LOG.info("creating; entries = " + length);
// create a file with length entries
SequenceFile.Writer writer =
SequenceFile.createWriter(fs, conf, file, IntWritable.class, BytesWritable.class);
try {
for (int i = 0; i < length; i++) {
IntWritable key = new IntWritable(i);
byte[] data = new byte[random.nextInt(10)];
random.nextBytes(data);
BytesWritable value = new BytesWritable(data);
writer.append(key, value);
}
} finally {
writer.close();
}
// try splitting the file in a variety of sizes
InputFormat<IntWritable, BytesWritable> format =
new SequenceFileInputFormat<IntWritable, BytesWritable>();
IntWritable key = new IntWritable();
BytesWritable value = new BytesWritable();
for (int i = 0; i < 3; i++) {
int numSplits = random.nextInt(MAX_LENGTH / (SequenceFile.SYNC_INTERVAL / 20)) + 1;
// LOG.info("splitting: requesting = " + numSplits);
InputSplit[] splits = format.getSplits(job, numSplits);
// LOG.info("splitting: got = " + splits.length);
// check each split
BitSet bits = new BitSet(length);
for (int j = 0; j < splits.length; j++) {
RecordReader<IntWritable, BytesWritable> reader =
format.getRecordReader(splits[j], job, reporter);
try {
int count = 0;
while (reader.next(key, value)) {
// if (bits.get(key.get())) {
// LOG.info("splits["+j+"]="+splits[j]+" : " +
// key.get());
// LOG.info("@"+reader.getPos());
// }
assertFalse("Key in multiple partitions.", bits.get(key.get()));
bits.set(key.get());
count++;
}
// LOG.info("splits["+j+"]="+splits[j]+" count=" +
// count);
} finally {
reader.close();
}
}
assertEquals("Some keys in no partition.", length, bits.cardinality());
}
}
}
public int run(String[] args) throws IOException, ClassNotFoundException, InterruptedException {
if (args.length != 2) {
System.out.println("Usage: FeatureMatching ID <inputName.jpeg/inputName.jpg>");
System.exit(1);
}
SimpleDateFormat sdf = new SimpleDateFormat("", Locale.US);
sdf.applyPattern("yyyy-MM-dd_HH-mm-ss");
String time = sdf.format(new Date());
Job job = Job.getInstance();
ID = "/" + args[0];
String filename = args[1];
filename = filename.toLowerCase();
System.out.println("current filename:" + filename);
// Detect illegal username (if the username dir doesn't exist)
File userPath = new File(LOCAL_USER_DIR + ID);
if (!userPath.exists()) {
System.out.println("Unauthorized username!!!\nExiting......");
System.exit(1);
}
// Preprocess the input image.jpg from local dir: /local.../user/ID/input/image.jpg
// Save the features to local dir: hdfs://.../user/ID/input/image.jpg
extractQueryFeatures2HDFS(filename, job);
// Add the feature file to the hdfs cache
String featureFileName = filename.substring(0, filename.lastIndexOf(".")) + ".json";
// job.addCacheFile(new Path(HDFS_HOME + USER + ID + INPUT + "/" +
// featureFileName).toUri());
job.getConfiguration()
.set("featureFilePath", HDFS_HOME + USER + ID + INPUT + "/" + featureFileName);
// Check the file type. Only support jpeg/jpg type images
String type = filename.substring(args[1].lastIndexOf("."));
if (!(type.equals(".jpg") || type.equals(".jpeg"))) {
System.out.println("Image type not supported!!!\nExiting");
System.exit(1);
}
// Input: hdfs://.../features/
// The feature dir is a location of all features extracted from the database
String inputPathStr = HDFS_HOME + FEATURES;
// Output: hdfs://.../user/ID/output/
String outputPathStr = HDFS_HOME + USER + ID + OUTPUT + "/" + time;
job.setInputFormatClass(KeyValueTextInputFormat.class);
// job.setOutputFormatClass(TextOutputFormat.class);
// Get the lists of all feature files: /.../features/data/part-*
FileSystem fs = FileSystem.get(job.getConfiguration());
FileStatus[] statuses = fs.listStatus(new Path(inputPathStr));
StringBuffer sb = new StringBuffer();
for (FileStatus fileStatus : statuses) {
sb.append(fileStatus.getPath() + ",");
}
sb.deleteCharAt(sb.lastIndexOf(","));
job.setJarByClass(FeatureMatching.class);
job.setMapperClass(FeatureMatchMapper.class);
job.setReducerClass(FeatureMatchReducer.class);
// only need one reducer to collect the result
job.setNumReduceTasks(1);
job.setMapOutputKeyClass(IntWritable.class);
job.setMapOutputValueClass(Text.class);
job.setOutputKeyClass(Text.class);
job.setOutputValueClass(DoubleWritable.class);
// Input a directory, so need the recursive input
FileInputFormat.setInputDirRecursive(job, true);
// Set the PathFilter, to omit _SUCCESS files
// (This is not working correctly, as the PathFilter class is an interface rather than a class.
// But the 2nd arg asks me to extend the PathFilter)
// FileInputFormat.setInputPathFilter(job, MyPathFilter.class);
//
// FileInputFormat.setInputPaths(job, new Path(inputPathStr));
FileInputFormat.setInputPaths(job, sb.toString());
FileOutputFormat.setOutputPath(job, new Path(outputPathStr));
boolean success = job.waitForCompletion(true);
return success ? 0 : 1;
}