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));
}
@Override
public int run(String[] arg0) throws Exception {
// config a job and start it
Configuration conf = getConf();
Job job = new Job(conf, "Index construction..");
job.setJarByClass(IndexConstructorDriver.class);
job.setMapperClass(IndexConstructorMapper.class);
job.setReducerClass(IndexConstructorReducer.class);
job.setOutputKeyClass(Text.class);
job.setOutputValueClass(InvertedListWritable.class);
job.setMapOutputKeyClass(Text.class);
job.setMapOutputValueClass(Text.class);
job.setInputFormatClass(SequenceFileInputFormat.class);
job.setOutputFormatClass(SequenceFileOutputFormat.class);
// can add the dir by the config
FileSystem fs = FileSystem.get(conf);
String workdir = conf.get("org.joy.crawler.dir", "crawler/");
fs.delete(new Path(workdir + "indexOutput/"), true);
FileInputFormat.addInputPath(job, new Path(workdir + "content/"));
FileOutputFormat.setOutputPath(job, new Path(workdir + "indexOutput/"));
System.out.println(
"indexer starts to work, it begins to construct the index, please wait ...\n");
return job.waitForCompletion(true) ? 0 : 1;
}
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()
+ ")");
}
}