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));
}
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;
}
void checkFormat(Job job) throws Exception {
TaskAttemptContext attemptContext =
new TaskAttemptContext(job.getConfiguration(), new TaskAttemptID("123", 0, false, 1, 2));
MyClassMessagePackBase64LineInputFormat format = new MyClassMessagePackBase64LineInputFormat();
FileInputFormat.setInputPaths(job, workDir);
List<InputSplit> splits = format.getSplits(job);
for (int j = 0; j < splits.size(); j++) {
RecordReader<LongWritable, MyClassWritable> reader =
format.createRecordReader(splits.get(j), attemptContext);
reader.initialize(splits.get(j), attemptContext);
int count = 0;
try {
while (reader.nextKeyValue()) {
LongWritable key = reader.getCurrentKey();
MyClassWritable val = reader.getCurrentValue();
MyClass mc = val.get();
assertEquals(mc.v, count);
assertEquals(mc.s, Integer.toString(count));
count++;
}
} finally {
reader.close();
}
}
}
private static IntWritable deduceInputFile(JobConf job) {
Path[] inputPaths = FileInputFormat.getInputPaths(job);
Path inputFile = new Path(job.get("map.input.file"));
// value == one for sort-input; value == two for sort-output
return (inputFile.getParent().equals(inputPaths[0])) ? sortInput : sortOutput;
}
private Job configureJob(Path secretsPath, Path saltFilePath, Path inputPath, Path outputPath)
throws Exception {
Job job = Job.getInstance(getConf());
FileInputFormat.setInputPaths(job, inputPath);
FileOutputFormat.setOutputPath(job, outputPath);
job.getConfiguration().set(ObfuscateMapper.SECRET_WORDS_FILE_KEY, secretsPath.toString());
job.getConfiguration().set(ObfuscateMapper.SALT_FILE_KEY, saltFilePath.toString());
job.setInputFormatClass(TextInputFormat.class);
job.setMapperClass(ObfuscateMapper.class);
job.setNumReduceTasks(0);
job.setJarByClass(getClass());
FileSystem.get(outputPath.toUri(), getConf()).delete(outputPath, true);
return job;
}
public int run(String[] args) throws Exception {
Path tempDir = new Path("/user/akhfa/temp");
Configuration conf = new Configuration();
Job job = Job.getInstance(conf, "word count");
job.setJarByClass(AuthorCounter.class);
job.setMapperClass(TokenizerMapper.class);
job.setCombinerClass(IntSumReducer.class);
job.setReducerClass(IntSumReducer.class);
job.setOutputKeyClass(Text.class);
job.setOutputValueClass(LongWritable.class);
job.setOutputFormatClass(SequenceFileOutputFormat.class);
FileInputFormat.addInputPath(job, new Path(args[0]));
FileOutputFormat.setOutputPath(job, tempDir);
System.exit(job.waitForCompletion(true) ? 0 : 1);
return 0;
}
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());
}
}
}
static void checkRecords(
Configuration defaults, int noMaps, int noReduces, Path sortInput, Path sortOutput)
throws IOException {
JobConf jobConf = new JobConf(defaults, RecordChecker.class);
jobConf.setJobName("sortvalidate-record-checker");
jobConf.setInputFormat(SequenceFileInputFormat.class);
jobConf.setOutputFormat(SequenceFileOutputFormat.class);
jobConf.setOutputKeyClass(BytesWritable.class);
jobConf.setOutputValueClass(IntWritable.class);
jobConf.setMapperClass(Map.class);
jobConf.setReducerClass(Reduce.class);
JobClient client = new JobClient(jobConf);
ClusterStatus cluster = client.getClusterStatus();
if (noMaps == -1) {
noMaps = cluster.getTaskTrackers() * jobConf.getInt("test.sortvalidate.maps_per_host", 10);
}
if (noReduces == -1) {
noReduces = (int) (cluster.getMaxReduceTasks() * 0.9);
String sortReduces = jobConf.get("test.sortvalidate.reduces_per_host");
if (sortReduces != null) {
noReduces = cluster.getTaskTrackers() * Integer.parseInt(sortReduces);
}
}
jobConf.setNumMapTasks(noMaps);
jobConf.setNumReduceTasks(noReduces);
FileInputFormat.setInputPaths(jobConf, sortInput);
FileInputFormat.addInputPath(jobConf, sortOutput);
Path outputPath = new Path("/tmp/sortvalidate/recordchecker");
FileSystem fs = FileSystem.get(defaults);
if (fs.exists(outputPath)) {
fs.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.RecordChecker: Running on "
+ cluster.getTaskTrackers()
+ " nodes to validate sort from "
+ inputPaths[0]
+ ", "
+ inputPaths[1]
+ " into "
+ FileOutputFormat.getOutputPath(jobConf)
+ " with "
+ noReduces
+ " reduces.");
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.");
}
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()
+ ")");
}
}