public void testComplexNameWithRegex() throws Exception {
OutputStream os = getFileSystem().create(new Path(getInputDir(), "text.txt"));
Writer wr = new OutputStreamWriter(os);
wr.write("b a\n");
wr.close();
JobConf conf = createJobConf();
conf.setJobName("name \\Evalue]");
conf.setInputFormat(TextInputFormat.class);
conf.setOutputKeyClass(LongWritable.class);
conf.setOutputValueClass(Text.class);
conf.setMapperClass(IdentityMapper.class);
FileInputFormat.setInputPaths(conf, getInputDir());
FileOutputFormat.setOutputPath(conf, getOutputDir());
JobClient.runJob(conf);
Path[] outputFiles =
FileUtil.stat2Paths(getFileSystem().listStatus(getOutputDir(), new OutputLogFilter()));
assertEquals(1, outputFiles.length);
InputStream is = getFileSystem().open(outputFiles[0]);
BufferedReader reader = new BufferedReader(new InputStreamReader(is));
assertEquals("0\tb a", reader.readLine());
assertNull(reader.readLine());
reader.close();
}
/**
* 입력으로 선택한 경로를 지정한 목적 경로로 이동한다.
*
* @param source 이동할 경로
* @param target 이동할 위치
* @param fs Hadoop FileSystem
*/
public static void move(String source, String target, FileSystem fs) throws Exception {
Path srcPath = new Path(source);
Path[] srcs = FileUtil.stat2Paths(fs.globStatus(srcPath), srcPath);
Path dst = new Path(target);
if (srcs.length > 1 && !fs.getFileStatus(dst).isDir()) {
throw new FileSystemException(
"When moving multiple files, destination should be a directory.");
}
for (int i = 0; i < srcs.length; i++) {
if (!fs.rename(srcs[i], dst)) {
FileStatus srcFstatus = null;
FileStatus dstFstatus = null;
try {
srcFstatus = fs.getFileStatus(srcs[i]);
} catch (FileNotFoundException e) {
throw new FileNotFoundException(srcs[i] + ": No such file or directory");
}
try {
dstFstatus = fs.getFileStatus(dst);
} catch (IOException e) {
// Nothing
}
if ((srcFstatus != null) && (dstFstatus != null)) {
if (srcFstatus.isDir() && !dstFstatus.isDir()) {
throw new FileSystemException(
"cannot overwrite non directory " + dst + " with directory " + srcs[i]);
}
}
throw new FileSystemException("Failed to rename " + srcs[i] + " to " + dst);
}
}
}
public static boolean globDelete(final FileSystem fs, final String path, final boolean recursive)
throws IOException {
boolean deleted = false;
for (final Path p : FileUtil.stat2Paths(fs.globStatus(new Path(path)))) {
fs.delete(p, recursive);
deleted = true;
}
return deleted;
}
@Test(timeout = 30000)
public void testStat2Paths1() {
assertNull(FileUtil.stat2Paths(null));
FileStatus[] fileStatuses = new FileStatus[0];
Path[] paths = FileUtil.stat2Paths(fileStatuses);
assertEquals(0, paths.length);
Path path1 = new Path("file://foo");
Path path2 = new Path("file://moo");
fileStatuses =
new FileStatus[] {
new FileStatus(3, false, 0, 0, 0, path1), new FileStatus(3, false, 0, 0, 0, path2)
};
paths = FileUtil.stat2Paths(fileStatuses);
assertEquals(2, paths.length);
assertEquals(paths[0], path1);
assertEquals(paths[1], path2);
}
public static Path getOutputsFinalJob(final FileSystem fs, final String output)
throws IOException {
int largest = -1;
for (final Path path : FileUtil.stat2Paths(fs.listStatus(new Path(output)))) {
final String[] name = path.getName().split(DASH);
if (name.length == 2 && name[0].equals(Tokens.JOB)) {
if (Integer.valueOf(name[1]) > largest) largest = Integer.valueOf(name[1]);
}
}
if (largest == -1) return new Path(output);
else return new Path(output + "/" + Tokens.JOB + "-" + largest);
}
public static void fileTreeRecursion(URI uri, Configuration conf, FileSystem fs)
throws IOException {
Path current = new Path(uri);
if (fs.isFile(current)) {
visit(current, fs);
} else {
FileStatus[] status = fs.listStatus(current);
Path[] paths = FileUtil.stat2Paths(status);
for (Path p : paths) {
fileTreeRecursion(p.toUri(), conf, fs);
}
}
}
/** Open the output generated by this format. */
public static MapFile.Reader[] getReaders(Path dir, Configuration conf) throws IOException {
FileSystem fs = dir.getFileSystem(conf);
Path[] names = FileUtil.stat2Paths(fs.listStatus(dir));
// sort names, so that hash partitioning works
Arrays.sort(names);
MapFile.Reader[] parts = new MapFile.Reader[names.length];
for (int i = 0; i < names.length; i++) {
parts[i] = new MapFile.Reader(fs, names[i].toString(), conf);
}
return parts;
}
@Test(timeout = 30000)
public void testStat2Paths2() {
Path defaultPath = new Path("file://default");
Path[] paths = FileUtil.stat2Paths(null, defaultPath);
assertEquals(1, paths.length);
assertEquals(defaultPath, paths[0]);
paths = FileUtil.stat2Paths(null, null);
assertTrue(paths != null);
assertEquals(1, paths.length);
assertEquals(null, paths[0]);
Path path1 = new Path("file://foo");
Path path2 = new Path("file://moo");
FileStatus[] fileStatuses =
new FileStatus[] {
new FileStatus(3, false, 0, 0, 0, path1), new FileStatus(3, false, 0, 0, 0, path2)
};
paths = FileUtil.stat2Paths(fileStatuses, defaultPath);
assertEquals(2, paths.length);
assertEquals(paths[0], path1);
assertEquals(paths[1], path2);
}
public static void main(String[] args) throws Exception {
String uri = args[0];
Configuration conf = new Configuration();
FileSystem fs = FileSystem.get(URI.create(uri), conf);
Path[] paths = new Path[args.length];
for (int i = 0; i < paths.length; i++) {
paths[i] = new Path(args[i]);
}
FileStatus[] status = fs.listStatus(paths);
Path[] listedPaths = FileUtil.stat2Paths(status);
for (Path p : listedPaths) {
System.out.println(p);
}
}
public static void decompressPath(
final FileSystem fs,
final String in,
final String out,
final String compressedFileSuffix,
final boolean deletePrevious)
throws IOException {
final Path inPath = new Path(in);
if (fs.isFile(inPath)) HDFSTools.decompressFile(fs, in, out, deletePrevious);
else {
final Path outPath = new Path(out);
if (!fs.exists(outPath)) fs.mkdirs(outPath);
for (final Path path : FileUtil.stat2Paths(fs.globStatus(new Path(in + FOWARD_ASTERISK)))) {
if (path.getName().endsWith(compressedFileSuffix))
HDFSTools.decompressFile(
fs,
path.toString(),
outPath.toString() + FOWARD_SLASH + path.getName().split("\\.")[0],
deletePrevious);
}
}
}
@Test
public void mrRun() throws Exception {
FileSystem fs = dfsCluster.getFileSystem();
Path inDir = fs.makeQualified(new Path("/user/testing/testMapperReducer/input"));
fs.delete(inDir, true);
String DATADIR = "/user/testing/testMapperReducer/data";
Path dataDir = fs.makeQualified(new Path(DATADIR));
fs.delete(dataDir, true);
Path outDir = fs.makeQualified(new Path("/user/testing/testMapperReducer/output"));
fs.delete(outDir, true);
assertTrue(fs.mkdirs(inDir));
Path INPATH = new Path(inDir, "input.txt");
OutputStream os = fs.create(INPATH);
Writer wr = new OutputStreamWriter(os, StandardCharsets.UTF_8);
wr.write(DATADIR + "/" + inputAvroFile);
wr.close();
assertTrue(fs.mkdirs(dataDir));
fs.copyFromLocalFile(new Path(DOCUMENTS_DIR, inputAvroFile), dataDir);
JobConf jobConf = getJobConf();
jobConf.set("jobclient.output.filter", "ALL");
if (ENABLE_LOCAL_JOB_RUNNER) { // enable Hadoop LocalJobRunner; this enables to run in debugger
// and set breakpoints
jobConf.set("mapred.job.tracker", "local");
}
jobConf.setMaxMapAttempts(1);
jobConf.setMaxReduceAttempts(1);
jobConf.setJar(SEARCH_ARCHIVES_JAR);
int shards = 2;
int maxReducers = Integer.MAX_VALUE;
if (ENABLE_LOCAL_JOB_RUNNER) {
// local job runner has a couple of limitations: only one reducer is supported and the
// DistributedCache doesn't work.
// see http://blog.cloudera.com/blog/2009/07/advice-on-qa-testing-your-mapreduce-jobs/
maxReducers = 1;
shards = 1;
}
String[] args =
new String[] {
"--morphline-file=" + tempDir + "/test-morphlines/solrCellDocumentTypes.conf",
"--morphline-id=morphline1",
"--solr-home-dir=" + MINIMR_CONF_DIR.getAbsolutePath(),
"--output-dir=" + outDir.toString(),
"--shards=" + shards,
"--verbose",
numRuns % 2 == 0 ? "--input-list=" + INPATH.toString() : dataDir.toString(),
numRuns % 3 == 0
? "--reducers=" + shards
: (numRuns % 3 == 1 ? "--reducers=-1" : "--reducers=" + Math.min(8, maxReducers))
};
if (numRuns % 3 == 2) {
args = concat(args, new String[] {"--fanout=2"});
}
if (numRuns == 0) {
// force (slow) MapReduce based randomization to get coverage for that as well
args =
concat(
new String[] {"-D", MapReduceIndexerTool.MAIN_MEMORY_RANDOMIZATION_THRESHOLD + "=-1"},
args);
}
MapReduceIndexerTool tool = createTool();
int res = ToolRunner.run(jobConf, tool, args);
assertEquals(0, res);
Job job = tool.job;
assertTrue(job.isComplete());
assertTrue(job.isSuccessful());
if (numRuns % 3 != 2) {
// Only run this check if mtree merge is disabled.
// With mtree merge enabled the BatchWriter counters aren't available anymore because
// variable "job" now refers to the merge job rather than the indexing job
assertEquals(
"Invalid counter "
+ SolrRecordWriter.class.getName()
+ "."
+ SolrCounters.DOCUMENTS_WRITTEN,
count,
job.getCounters()
.findCounter(SolrCounters.class.getName(), SolrCounters.DOCUMENTS_WRITTEN.toString())
.getValue());
}
// Check the output is as expected
outDir = new Path(outDir, MapReduceIndexerTool.RESULTS_DIR);
Path[] outputFiles = FileUtil.stat2Paths(fs.listStatus(outDir));
System.out.println("outputfiles:" + Arrays.toString(outputFiles));
UtilsForTests.validateSolrServerDocumentCount(MINIMR_CONF_DIR, fs, outDir, count, shards);
// run again with --dryrun mode:
tool = createTool();
args = concat(args, new String[] {"--dry-run"});
res = ToolRunner.run(jobConf, tool, args);
assertEquals(0, res);
numRuns++;
}
@Override
public List<InputSplit> getSplits(JobContext job) throws IOException {
long minSizeNode = 0;
long minSizeRack = 0;
long maxSize = 0;
Configuration conf = job.getConfiguration();
// the values specified by setxxxSplitSize() takes precedence over the
// values that might have been specified in the config
if (minSplitSizeNode != 0) {
minSizeNode = minSplitSizeNode;
} else {
minSizeNode = conf.getLong("mapred.min.split.size.per.node", 0);
}
if (minSplitSizeRack != 0) {
minSizeRack = minSplitSizeRack;
} else {
minSizeRack = conf.getLong("mapred.min.split.size.per.rack", 0);
}
if (maxSplitSize != 0) {
maxSize = maxSplitSize;
} else {
maxSize = conf.getLong("mapred.max.split.size", 0);
}
if (minSizeNode != 0 && maxSize != 0 && minSizeNode > maxSize) {
throw new IOException(
"Minimum split size pernode "
+ minSizeNode
+ " cannot be larger than maximum split size "
+ maxSize);
}
if (minSizeRack != 0 && maxSize != 0 && minSizeRack > maxSize) {
throw new IOException(
"Minimum split size per rack"
+ minSizeRack
+ " cannot be larger than maximum split size "
+ maxSize);
}
if (minSizeRack != 0 && minSizeNode > minSizeRack) {
throw new IOException(
"Minimum split size per node"
+ minSizeNode
+ " cannot be smaller than minimum split "
+ "size per rack "
+ minSizeRack);
}
// all the files in input set
Path[] paths = FileUtil.stat2Paths(listStatus(job).toArray(new FileStatus[0]));
List<InputSplit> splits = new ArrayList<InputSplit>();
if (paths.length == 0) {
return splits;
}
// In one single iteration, process all the paths in a single pool.
// Processing one pool at a time ensures that a split contains paths
// from a single pool only.
for (MultiPathFilter onepool : pools) {
ArrayList<Path> myPaths = new ArrayList<Path>();
// pick one input path. If it matches all the filters in a pool,
// add it to the output set
for (int i = 0; i < paths.length; i++) {
if (paths[i] == null) { // already processed
continue;
}
Path p = new Path(paths[i].toUri().getPath());
if (onepool.accept(p)) {
myPaths.add(paths[i]); // add it to my output set
paths[i] = null; // already processed
}
}
// create splits for all files in this pool.
getMoreSplits(
conf,
myPaths.toArray(new Path[myPaths.size()]),
maxSize,
minSizeNode,
minSizeRack,
splits);
}
// Finally, process all paths that do not belong to any pool.
ArrayList<Path> myPaths = new ArrayList<Path>();
for (int i = 0; i < paths.length; i++) {
if (paths[i] == null) { // already processed
continue;
}
myPaths.add(paths[i]);
}
// create splits for all files that are not in any pool.
getMoreSplits(
conf, myPaths.toArray(new Path[myPaths.size()]), maxSize, minSizeNode, minSizeRack, splits);
// free up rackToNodes map
rackToNodes.clear();
return splits;
}
public void testSymLink() {
try {
boolean mayExit = false;
MiniMRCluster mr = null;
MiniDFSCluster dfs = null;
try {
Configuration conf = new Configuration();
dfs = new MiniDFSCluster(conf, 1, true, null);
FileSystem fileSys = dfs.getFileSystem();
String namenode = fileSys.getUri().toString();
mr = new MiniMRCluster(1, namenode, 3);
// During tests, the default Configuration will use a local mapred
// So don't specify -config or -cluster
String strJobtracker = "mapred.job.tracker=" + "localhost:" + mr.getJobTrackerPort();
String strNamenode = "fs.default.name=" + namenode;
String argv[] =
new String[] {
"-input", INPUT_FILE,
"-output", OUTPUT_DIR,
"-mapper", map,
"-reducer", reduce,
// "-verbose",
// "-jobconf", "stream.debug=set"
"-jobconf", strNamenode,
"-jobconf", strJobtracker,
"-jobconf", "stream.tmpdir=" + System.getProperty("test.build.data", "/tmp"),
"-jobconf",
"mapred.child.java.opts=-Dcontrib.name="
+ System.getProperty("contrib.name")
+ " "
+ "-Dbuild.test="
+ System.getProperty("build.test")
+ " "
+ conf.get("mapred.child.java.opts", ""),
"-cacheFile", fileSys.getUri() + CACHE_FILE + "#testlink"
};
fileSys.delete(new Path(OUTPUT_DIR), true);
DataOutputStream file = fileSys.create(new Path(INPUT_FILE));
file.writeBytes(mapString);
file.close();
file = fileSys.create(new Path(CACHE_FILE));
file.writeBytes(cacheString);
file.close();
job = new StreamJob(argv, mayExit);
job.go();
fileSys = dfs.getFileSystem();
String line = null;
Path[] fileList =
FileUtil.stat2Paths(fileSys.listStatus(new Path(OUTPUT_DIR), new OutputLogFilter()));
for (int i = 0; i < fileList.length; i++) {
System.out.println(fileList[i].toString());
BufferedReader bread =
new BufferedReader(new InputStreamReader(fileSys.open(fileList[i])));
line = bread.readLine();
System.out.println(line);
}
assertEquals(cacheString + "\t", line);
} finally {
if (dfs != null) {
dfs.shutdown();
}
if (mr != null) {
mr.shutdown();
}
}
} catch (Exception e) {
failTrace(e);
}
}
public void doTestTextBatchAppend() throws Exception {
LOG.debug("Starting...");
final long rollCount = 10;
final long batchSize = 2;
final String fileName = "PageView";
String newPath = testPath + "/singleTextBucket";
int totalEvents = 0;
int i = 1, j = 1;
// clear the test directory
Configuration conf = new Configuration();
FileSystem fs = FileSystem.get(conf);
Path dirPath = new Path(newPath);
fs.delete(dirPath, true);
fs.mkdirs(dirPath);
Context context = new Context();
context.put("hdfs.path", newPath);
context.put("hdfs.rollCount", String.valueOf(rollCount));
context.put("hdfs.batchSize", String.valueOf(batchSize));
context.put("hdfs.filePrefix", "pageview");
Channel channel = new MemoryChannel();
Configurables.configure(channel, context);
sink.setChannel(channel);
sink.start();
Calendar eventDate = Calendar.getInstance();
Date currentDate = new Date();
Map<String, String> header = new HashMap<String, String>();
header.put("topic", "PageView");
List<String> bodies = Lists.newArrayList();
// 将测试的事件推入到通道中
for (i = 1; i <= (rollCount * 10) / batchSize; i++) {
Transaction txn = channel.getTransaction();
txn.begin();
for (j = 1; j <= batchSize; j++) {
header.put("timestamp", String.valueOf(currentDate.getTime()));
Event event = new SimpleEvent();
eventDate.clear();
eventDate.set(2014, i, i, i, 0);
String body = "Test." + i + "." + j;
event.setHeaders(header);
event.setBody(body.getBytes());
bodies.add(body);
channel.put(event);
totalEvents++;
}
txn.commit();
txn.close();
// execute sink to process the events
sink.process();
}
sink.stop();
FileStatus[] dirStat = fs.listStatus(dirPath);
Path fList[] = FileUtil.stat2Paths(dirStat);
long expectedFiles = totalEvents / rollCount;
if (totalEvents % rollCount > 0) {
expectedFiles++;
}
Assert.assertEquals(
"num files wrong, found: " + Lists.newArrayList(fList), expectedFiles, fList.length);
// 检查所有写入文件的内容
verifyOutputTextFiles(fs, conf, dirPath.toUri().getPath(), fileName, bodies);
}