private static void finalize(
Configuration conf, JobConf jobconf, final Path destPath, String presevedAttributes)
throws IOException {
if (presevedAttributes == null) {
return;
}
EnumSet<FileAttribute> preseved = FileAttribute.parse(presevedAttributes);
if (!preseved.contains(FileAttribute.USER)
&& !preseved.contains(FileAttribute.GROUP)
&& !preseved.contains(FileAttribute.PERMISSION)) {
return;
}
FileSystem dstfs = destPath.getFileSystem(conf);
Path dstdirlist = new Path(jobconf.get(DST_DIR_LIST_LABEL));
SequenceFile.Reader in = null;
try {
in = new SequenceFile.Reader(dstdirlist.getFileSystem(jobconf), dstdirlist, jobconf);
Text dsttext = new Text();
FilePair pair = new FilePair();
for (; in.next(dsttext, pair); ) {
Path absdst = new Path(destPath, pair.output);
updatePermissions(pair.input, dstfs.getFileStatus(absdst), preseved, dstfs);
}
} finally {
checkAndClose(in);
}
}
// copy source to target with verification
static void copyAndVerify(Path source, Path target, CopyOption... options) throws IOException {
Path result = copy(source, target, options);
assertTrue(result == target);
// get attributes of source and target file to verify copy
boolean followLinks = true;
LinkOption[] linkOptions = new LinkOption[0];
boolean copyAttributes = false;
for (CopyOption opt : options) {
if (opt == NOFOLLOW_LINKS) {
followLinks = false;
linkOptions = new LinkOption[] {NOFOLLOW_LINKS};
}
if (opt == COPY_ATTRIBUTES) copyAttributes = true;
}
BasicFileAttributes basicAttributes =
readAttributes(source, BasicFileAttributes.class, linkOptions);
// check hash if regular file
if (basicAttributes.isRegularFile()) assertTrue(computeHash(source) == computeHash(target));
// check link target if symbolic link
if (basicAttributes.isSymbolicLink())
assert (readSymbolicLink(source).equals(readSymbolicLink(target)));
// check that attributes are copied
if (copyAttributes && followLinks) {
checkBasicAttributes(
basicAttributes, readAttributes(source, BasicFileAttributes.class, linkOptions));
// verify other attributes when same provider
if (source.getFileSystem().provider() == target.getFileSystem().provider()) {
// check POSIX attributes are copied
String os = System.getProperty("os.name");
if ((os.equals("SunOS") || os.equals("Linux")) && testPosixAttributes) {
checkPosixAttributes(
readAttributes(source, PosixFileAttributes.class, linkOptions),
readAttributes(target, PosixFileAttributes.class, linkOptions));
}
// check DOS attributes are copied
if (os.startsWith("Windows")) {
checkDosAttributes(
readAttributes(source, DosFileAttributes.class, linkOptions),
readAttributes(target, DosFileAttributes.class, linkOptions));
}
// check named attributes are copied
if (followLinks
&& getFileStore(source).supportsFileAttributeView("xattr")
&& getFileStore(target).supportsFileAttributeView("xattr")) {
checkUserDefinedFileAttributes(
readUserDefinedFileAttributes(source), readUserDefinedFileAttributes(target));
}
}
}
}
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();
}
/*
* Fetch a file that is in a Hadoop file system. Return a local File.
* Interruptible.
*/
private File hdfsFetch(Path fromPath, Reporter reporter)
throws IOException, InterruptedException {
UUID uniqueId = UUID.randomUUID();
File toFile = new File(tempDir, uniqueId.toString() + "/" + fromPath.getName());
File toDir = new File(toFile.getParent());
if (toDir.exists()) {
FileUtils.deleteDirectory(toDir);
}
toDir.mkdirs();
Path toPath = new Path(toFile.getCanonicalPath());
FileSystem fS = fromPath.getFileSystem(hadoopConf);
FileSystem tofS = FileSystem.getLocal(hadoopConf);
Throttler throttler = new Throttler((double) bytesPerSecThrottle);
try {
for (FileStatus fStatus : fS.globStatus(fromPath)) {
log.info("Copying " + fStatus.getPath() + " to " + toPath);
long bytesSoFar = 0;
FSDataInputStream iS = fS.open(fStatus.getPath());
FSDataOutputStream oS = tofS.create(toPath);
byte[] buffer = new byte[downloadBufferSize];
int nRead;
while ((nRead = iS.read(buffer, 0, buffer.length)) != -1) {
// Needed to being able to be interrupted at any moment.
if (Thread.interrupted()) {
iS.close();
oS.close();
cleanDirNoExceptions(toDir);
throw new InterruptedException();
}
bytesSoFar += nRead;
oS.write(buffer, 0, nRead);
throttler.incrementAndThrottle(nRead);
if (bytesSoFar >= bytesToReportProgress) {
reporter.progress(bytesSoFar);
bytesSoFar = 0l;
}
}
if (reporter != null) {
reporter.progress(bytesSoFar);
}
oS.close();
iS.close();
}
return toDir;
} catch (ClosedByInterruptException e) {
// This can be thrown by the method read.
cleanDirNoExceptions(toDir);
throw new InterruptedIOException();
}
}
/**
* 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);
}
/**
* 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;
}
/** Sanity check for srcPath */
private static void checkSrcPath(Configuration conf, List<Path> srcPaths) throws IOException {
List<IOException> rslt = new ArrayList<IOException>();
for (Path p : srcPaths) {
FileSystem fs = p.getFileSystem(conf);
if (!fs.exists(p)) {
rslt.add(new IOException("Input source " + p + " does not exist."));
}
}
if (!rslt.isEmpty()) {
throw new InvalidInputException(rslt);
}
}
/**
* Produce splits such that each is no greater than the quotient of the total size and the
* number of splits requested.
*
* @param job The handle to the JobConf object
* @param numSplits Number of splits requested
*/
public InputSplit[] getSplits(JobConf job, int numSplits) throws IOException {
int cnfiles = job.getInt(SRC_COUNT_LABEL, -1);
long cbsize = job.getLong(TOTAL_SIZE_LABEL, -1);
String srcfilelist = job.get(SRC_LIST_LABEL, "");
if (cnfiles < 0 || cbsize < 0 || "".equals(srcfilelist)) {
throw new RuntimeException(
"Invalid metadata: #files("
+ cnfiles
+ ") total_size("
+ cbsize
+ ") listuri("
+ srcfilelist
+ ")");
}
Path src = new Path(srcfilelist);
FileSystem fs = src.getFileSystem(job);
FileStatus srcst = fs.getFileStatus(src);
ArrayList<FileSplit> splits = new ArrayList<FileSplit>(numSplits);
LongWritable key = new LongWritable();
FilePair value = new FilePair();
final long targetsize = cbsize / numSplits;
long pos = 0L;
long last = 0L;
long acc = 0L;
long cbrem = srcst.getLen();
SequenceFile.Reader sl = null;
try {
sl = new SequenceFile.Reader(fs, src, job);
for (; sl.next(key, value); last = sl.getPosition()) {
// if adding this split would put this split past the target size,
// cut the last split and put this next file in the next split.
if (acc + key.get() > targetsize && acc != 0) {
long splitsize = last - pos;
splits.add(new FileSplit(src, pos, splitsize, (String[]) null));
cbrem -= splitsize;
pos = last;
acc = 0L;
}
acc += key.get();
}
} finally {
checkAndClose(sl);
}
if (cbrem != 0) {
splits.add(new FileSplit(src, pos, cbrem, (String[]) null));
}
return splits.toArray(new FileSplit[splits.size()]);
}
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();
}
private static List<Path> fetchFileList(Configuration conf, Path srcList) throws IOException {
List<Path> result = new ArrayList<Path>();
FileSystem fs = srcList.getFileSystem(conf);
BufferedReader input = null;
try {
input = new BufferedReader(new InputStreamReader(fs.open(srcList)));
String line = input.readLine();
while (line != null) {
result.add(new Path(line));
line = input.readLine();
}
} finally {
checkAndClose(input);
}
return result;
}
/**
* Mapper configuration. Extracts source and destination file system, as well as top-level paths
* on source and destination directories. Gets the named file systems, to be used later in map.
*/
public void configure(JobConf job) {
destPath = new Path(job.get(DST_DIR_LABEL, "/"));
try {
destFileSys = destPath.getFileSystem(job);
} catch (IOException ex) {
throw new RuntimeException("Unable to get the named file system.", ex);
}
sizeBuf = job.getInt("copy.buf.size", 128 * 1024);
buffer = new byte[sizeBuf];
ignoreReadFailures = job.getBoolean(Options.IGNORE_READ_FAILURES.propertyname, false);
preserve_status = job.getBoolean(Options.PRESERVE_STATUS.propertyname, false);
if (preserve_status) {
preseved = FileAttribute.parse(job.get(PRESERVE_STATUS_LABEL));
}
update = job.getBoolean(Options.UPDATE.propertyname, false);
overwrite = !update && job.getBoolean(Options.OVERWRITE.propertyname, false);
this.job = job;
}
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);
}
}
/**
* 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));
}
/**
* 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));
}
/** Fully delete dir */
static void fullyDelete(String dir, Configuration conf) throws IOException {
if (dir != null) {
Path tmp = new Path(dir);
tmp.getFileSystem(conf).delete(tmp, true);
}
}
// move source to target with verification
static void moveAndVerify(Path source, Path target, CopyOption... options) throws IOException {
// read attributes before file is moved
BasicFileAttributes basicAttributes = null;
PosixFileAttributes posixAttributes = null;
DosFileAttributes dosAttributes = null;
Map<String, ByteBuffer> namedAttributes = null;
// get file attributes of source file
String os = System.getProperty("os.name");
if (os.startsWith("Windows")) {
dosAttributes = readAttributes(source, DosFileAttributes.class, NOFOLLOW_LINKS);
basicAttributes = dosAttributes;
} else {
posixAttributes = readAttributes(source, PosixFileAttributes.class, NOFOLLOW_LINKS);
basicAttributes = posixAttributes;
}
if (basicAttributes == null)
basicAttributes = readAttributes(source, BasicFileAttributes.class, NOFOLLOW_LINKS);
// hash file contents if regular file
int hash = (basicAttributes.isRegularFile()) ? computeHash(source) : 0;
// record link target if symbolic link
Path linkTarget = null;
if (basicAttributes.isSymbolicLink()) linkTarget = readSymbolicLink(source);
// read named attributes if available (and file is not a sym link)
if (!basicAttributes.isSymbolicLink()
&& getFileStore(source).supportsFileAttributeView("xattr")) {
namedAttributes = readUserDefinedFileAttributes(source);
}
// move file
Path result = move(source, target, options);
assertTrue(result == target);
// verify source does not exist
assertTrue(notExists(source));
// verify file contents
if (basicAttributes.isRegularFile()) {
if (computeHash(target) != hash)
throw new RuntimeException("Failed to verify move of regular file");
}
// verify link target
if (basicAttributes.isSymbolicLink()) {
if (!readSymbolicLink(target).equals(linkTarget))
throw new RuntimeException("Failed to verify move of symbolic link");
}
// verify basic attributes
checkBasicAttributes(
basicAttributes, readAttributes(target, BasicFileAttributes.class, NOFOLLOW_LINKS));
// verify other attributes when same provider
if (source.getFileSystem().provider() == target.getFileSystem().provider()) {
// verify POSIX attributes
if (posixAttributes != null && !basicAttributes.isSymbolicLink() && testPosixAttributes) {
checkPosixAttributes(
posixAttributes, readAttributes(target, PosixFileAttributes.class, NOFOLLOW_LINKS));
}
// verify DOS attributes
if (dosAttributes != null && !basicAttributes.isSymbolicLink()) {
DosFileAttributes attrs = readAttributes(target, DosFileAttributes.class, NOFOLLOW_LINKS);
checkDosAttributes(dosAttributes, attrs);
}
// verify named attributes
if (namedAttributes != null && getFileStore(target).supportsFileAttributeView("xattr")) {
checkUserDefinedFileAttributes(namedAttributes, readUserDefinedFileAttributes(target));
}
}
}
/**
* Initialize DFSCopyFileMapper specific job-configuration.
*
* @param conf : The dfs/mapred configuration.
* @param jobConf : The handle to the jobConf object to be initialized.
* @param args Arguments
*/
private static void setup(Configuration conf, JobConf jobConf, final Arguments args)
throws IOException {
jobConf.set(DST_DIR_LABEL, args.dst.toUri().toString());
// set boolean values
final boolean update = args.flags.contains(Options.UPDATE);
final boolean overwrite = !update && args.flags.contains(Options.OVERWRITE);
jobConf.setBoolean(Options.UPDATE.propertyname, update);
jobConf.setBoolean(Options.OVERWRITE.propertyname, overwrite);
jobConf.setBoolean(
Options.IGNORE_READ_FAILURES.propertyname,
args.flags.contains(Options.IGNORE_READ_FAILURES));
jobConf.setBoolean(
Options.PRESERVE_STATUS.propertyname, args.flags.contains(Options.PRESERVE_STATUS));
final String randomId = getRandomId();
JobClient jClient = new JobClient(jobConf);
Path jobDirectory = new Path(jClient.getSystemDir(), NAME + "_" + randomId);
jobConf.set(JOB_DIR_LABEL, jobDirectory.toString());
FileSystem dstfs = args.dst.getFileSystem(conf);
boolean dstExists = dstfs.exists(args.dst);
boolean dstIsDir = false;
if (dstExists) {
dstIsDir = dstfs.getFileStatus(args.dst).isDir();
}
// default logPath
Path logPath = args.log;
if (logPath == null) {
String filename = "_distcp_logs_" + randomId;
if (!dstExists || !dstIsDir) {
Path parent = args.dst.getParent();
if (!dstfs.exists(parent)) {
dstfs.mkdirs(parent);
}
logPath = new Path(parent, filename);
} else {
logPath = new Path(args.dst, filename);
}
}
FileOutputFormat.setOutputPath(jobConf, logPath);
// create src list, dst list
FileSystem jobfs = jobDirectory.getFileSystem(jobConf);
Path srcfilelist = new Path(jobDirectory, "_distcp_src_files");
jobConf.set(SRC_LIST_LABEL, srcfilelist.toString());
SequenceFile.Writer src_writer =
SequenceFile.createWriter(
jobfs,
jobConf,
srcfilelist,
LongWritable.class,
FilePair.class,
SequenceFile.CompressionType.NONE);
Path dstfilelist = new Path(jobDirectory, "_distcp_dst_files");
SequenceFile.Writer dst_writer =
SequenceFile.createWriter(
jobfs, jobConf, dstfilelist, Text.class, Text.class, SequenceFile.CompressionType.NONE);
Path dstdirlist = new Path(jobDirectory, "_distcp_dst_dirs");
jobConf.set(DST_DIR_LIST_LABEL, dstdirlist.toString());
SequenceFile.Writer dir_writer =
SequenceFile.createWriter(
jobfs,
jobConf,
dstdirlist,
Text.class,
FilePair.class,
SequenceFile.CompressionType.NONE);
// handle the case where the destination directory doesn't exist
// and we've only a single src directory OR we're updating/overwriting
// the contents of the destination directory.
final boolean special = (args.srcs.size() == 1 && !dstExists) || update || overwrite;
int srcCount = 0, cnsyncf = 0, dirsyn = 0;
long fileCount = 0L, byteCount = 0L, cbsyncs = 0L;
try {
for (Iterator<Path> srcItr = args.srcs.iterator(); srcItr.hasNext(); ) {
final Path src = srcItr.next();
FileSystem srcfs = src.getFileSystem(conf);
FileStatus srcfilestat = srcfs.getFileStatus(src);
Path root = special && srcfilestat.isDir() ? src : src.getParent();
if (srcfilestat.isDir()) {
++srcCount;
}
Stack<FileStatus> pathstack = new Stack<FileStatus>();
for (pathstack.push(srcfilestat); !pathstack.empty(); ) {
FileStatus cur = pathstack.pop();
FileStatus[] children = srcfs.listStatus(cur.getPath());
for (int i = 0; i < children.length; i++) {
boolean skipfile = false;
final FileStatus child = children[i];
final String dst = makeRelative(root, child.getPath());
++srcCount;
if (child.isDir()) {
pathstack.push(child);
} else {
// skip file if the src and the dst files are the same.
skipfile = update && sameFile(srcfs, child, dstfs, new Path(args.dst, dst));
// skip file if it exceed file limit or size limit
skipfile |=
fileCount == args.filelimit || byteCount + child.getLen() > args.sizelimit;
if (!skipfile) {
++fileCount;
byteCount += child.getLen();
if (LOG.isTraceEnabled()) {
LOG.trace("adding file " + child.getPath());
}
++cnsyncf;
cbsyncs += child.getLen();
if (cnsyncf > SYNC_FILE_MAX || cbsyncs > BYTES_PER_MAP) {
src_writer.sync();
dst_writer.sync();
cnsyncf = 0;
cbsyncs = 0L;
}
}
}
if (!skipfile) {
src_writer.append(
new LongWritable(child.isDir() ? 0 : child.getLen()), new FilePair(child, dst));
}
dst_writer.append(new Text(dst), new Text(child.getPath().toString()));
}
if (cur.isDir()) {
String dst = makeRelative(root, cur.getPath());
dir_writer.append(new Text(dst), new FilePair(cur, dst));
if (++dirsyn > SYNC_FILE_MAX) {
dirsyn = 0;
dir_writer.sync();
}
}
}
}
} finally {
checkAndClose(src_writer);
checkAndClose(dst_writer);
checkAndClose(dir_writer);
}
FileStatus dststatus = null;
try {
dststatus = dstfs.getFileStatus(args.dst);
} catch (FileNotFoundException fnfe) {
LOG.info(args.dst + " does not exist.");
}
// create dest path dir if copying > 1 file
if (dststatus == null) {
if (srcCount > 1 && !dstfs.mkdirs(args.dst)) {
throw new IOException("Failed to create" + args.dst);
}
}
final Path sorted = new Path(jobDirectory, "_distcp_sorted");
checkDuplication(jobfs, dstfilelist, sorted, conf);
if (dststatus != null && args.flags.contains(Options.DELETE)) {
deleteNonexisting(dstfs, dststatus, sorted, jobfs, jobDirectory, jobConf, conf);
}
Path tmpDir =
new Path(
(dstExists && !dstIsDir) || (!dstExists && srcCount == 1)
? args.dst.getParent()
: args.dst,
"_distcp_tmp_" + randomId);
jobConf.set(TMP_DIR_LABEL, tmpDir.toUri().toString());
LOG.info("srcCount=" + srcCount);
jobConf.setInt(SRC_COUNT_LABEL, srcCount);
jobConf.setLong(TOTAL_SIZE_LABEL, byteCount);
setMapCount(byteCount, jobConf);
}
static void checkRecords(Configuration defaults, Path sortInput, Path sortOutput)
throws IOException {
FileSystem inputfs = sortInput.getFileSystem(defaults);
FileSystem outputfs = sortOutput.getFileSystem(defaults);
FileSystem defaultfs = FileSystem.get(defaults);
JobConf jobConf = new JobConf(defaults, RecordStatsChecker.class);
jobConf.setJobName("sortvalidate-recordstats-checker");
int noSortReduceTasks = outputfs.listStatus(sortOutput, sortPathsFilter).length;
jobConf.setInt("sortvalidate.sort.reduce.tasks", noSortReduceTasks);
int noSortInputpaths = inputfs.listStatus(sortInput).length;
jobConf.setInputFormat(NonSplitableSequenceFileInputFormat.class);
jobConf.setOutputFormat(SequenceFileOutputFormat.class);
jobConf.setOutputKeyClass(IntWritable.class);
jobConf.setOutputValueClass(RecordStatsChecker.RecordStatsWritable.class);
jobConf.setMapperClass(Map.class);
jobConf.setCombinerClass(Reduce.class);
jobConf.setReducerClass(Reduce.class);
jobConf.setNumMapTasks(noSortReduceTasks);
jobConf.setNumReduceTasks(1);
FileInputFormat.setInputPaths(jobConf, sortInput);
FileInputFormat.addInputPath(jobConf, sortOutput);
Path outputPath = new Path("/tmp/sortvalidate/recordstatschecker");
if (defaultfs.exists(outputPath)) {
defaultfs.delete(outputPath, true);
}
FileOutputFormat.setOutputPath(jobConf, outputPath);
// Uncomment to run locally in a single process
// job_conf.set("mapred.job.tracker", "local");
Path[] inputPaths = FileInputFormat.getInputPaths(jobConf);
System.out.println(
"\nSortValidator.RecordStatsChecker: Validate sort "
+ "from "
+ inputPaths[0]
+ " ("
+ noSortInputpaths
+ " files), "
+ inputPaths[1]
+ " ("
+ noSortReduceTasks
+ " files) into "
+ FileOutputFormat.getOutputPath(jobConf)
+ " with 1 reducer.");
Date startTime = new Date();
System.out.println("Job started: " + startTime);
JobClient.runJob(jobConf);
Date end_time = new Date();
System.out.println("Job ended: " + end_time);
System.out.println(
"The job took " + (end_time.getTime() - startTime.getTime()) / 1000 + " seconds.");
// Check to ensure that the statistics of the
// framework's sort-input and sort-output match
SequenceFile.Reader stats =
new SequenceFile.Reader(defaultfs, new Path(outputPath, "part-00000"), defaults);
IntWritable k1 = new IntWritable();
IntWritable k2 = new IntWritable();
RecordStatsWritable v1 = new RecordStatsWritable();
RecordStatsWritable v2 = new RecordStatsWritable();
if (!stats.next(k1, v1)) {
throw new IOException("Failed to read record #1 from reduce's output");
}
if (!stats.next(k2, v2)) {
throw new IOException("Failed to read record #2 from reduce's output");
}
if ((v1.getBytes() != v2.getBytes())
|| (v1.getRecords() != v2.getRecords())
|| v1.getChecksum() != v2.getChecksum()) {
throw new IOException(
"("
+ v1.getBytes()
+ ", "
+ v1.getRecords()
+ ", "
+ v1.getChecksum()
+ ") v/s ("
+ v2.getBytes()
+ ", "
+ v2.getRecords()
+ ", "
+ v2.getChecksum()
+ ")");
}
}