@org.junit.Test
public void testMapper() throws Exception {
final ArcFileReader reader = new ArcFileReader();
Thread thread =
new Thread(
new Runnable() {
public void run() {
try {
while (reader.hasMoreItems()) {
ArcFileItem item = new ArcFileItem();
reader.getNextItem(item);
map(new Text(item.getUri()), item, null, null);
}
LOG.info("NO MORE ITEMS... BYE");
} catch (IOException e) {
LOG.error(StringUtils.stringifyException(e));
}
}
});
// run the thread ...
thread.start();
File file = new File("/Users/rana/Downloads/1213886083018_0.arc.gz");
ReadableByteChannel channel = Channels.newChannel(new FileInputStream(file));
try {
int totalBytesRead = 0;
for (; ; ) {
ByteBuffer buffer = ByteBuffer.allocate(ArcFileReader.DEFAULT_BLOCK_SIZE);
int bytesRead = channel.read(buffer);
LOG.info("Read " + bytesRead + " From File");
if (bytesRead == -1) {
reader.finished();
break;
} else {
buffer.flip();
totalBytesRead += buffer.remaining();
reader.available(buffer);
}
}
} finally {
channel.close();
}
// now wait for thread to die ...
LOG.info("Done Reading File.... Waiting for ArcFileThread to DIE");
thread.join();
LOG.info("Done Reading File.... ArcFileThread to DIED");
}
public static void main(String[] args) {
String accessKey = args[0];
String secretKey = args[1];
String paths[] = {
// "2008/06",
// "2008/07",
// "2008/08",
// "2008/09",
// "2008/10",
// "2008/11",
"2009"
};
for (int pathIndex = 0; pathIndex < paths.length; ++pathIndex) {
LOG.info("Processing Path:" + paths[pathIndex]);
JobConf job = new JobConf(S3GetMetdataJob.class);
Path tempDir =
new Path(
job.get("mapred.temp.dir", ".") + "/generate-temp-" + System.currentTimeMillis());
LOG.info("Output for Path:" + paths[pathIndex] + " is:" + tempDir);
System.out.println("Output Path is:" + tempDir);
job.setJobName("S3 To CrawlURLMetadata Job for Path:" + paths[pathIndex]);
// setup s3 properties
JetS3tARCSource.setMaxRetries(job, 1);
// set up S3 credentials ...
JetS3tARCSource.setAWSAccessKeyID(job, accessKey);
JetS3tARCSource.setAWSSecretAccessKey(job, secretKey);
ARCSplitCalculator.setFilesPerSplit(job, 25);
// set up arc reader properties
ArcFileReader.setIOTimeoutValue(30000);
// set input prefixes ...
JetS3tARCSource.setInputPrefixes(job, paths[pathIndex]);
// and S3 bucket name ...
JetS3tARCSource.setBucketName(job, "commoncrawl");
// and setup arc source for ArcInputFormat
ARCInputFormat.setARCSourceClass(job, JetS3tARCSource.class);
// and set up input format ...
job.setInputFormat(ARCInputFormat.class);
// set mapper ...
job.setMapRunnerClass(S3GetMetdataJob.class);
// setup reducer (identity in this case ... )
job.setReducerClass(IdentityReducer.class);
// standard output format ...
job.setOutputFormat(SequenceFileOutputFormat.class);
// set output path
FileOutputFormat.setOutputPath(job, tempDir);
// map output types
job.setMapOutputKeyClass(Text.class);
job.setMapOutputValueClass(CrawlURLMetadata.class);
// reduce output types
job.setOutputKeyClass(Text.class);
job.setOutputValueClass(CrawlURLMetadata.class);
// double the number of reducers ...
// job.setNumReduceTasks(job.getNumReduceTasks() * 2);
// run the job ...
try {
LOG.info("Starting Job:" + job.getJobName());
JobClient.runJob(job);
LOG.info("Finished Job:" + job.getJobName());
Path finalPath = new Path("jobout/" + paths[pathIndex] + "/result");
LOG.info("Copying Job Output to:" + finalPath);
FileSystem fs = FileSystem.get(job);
try {
fs.mkdirs(finalPath.getParent());
fs.rename(tempDir, finalPath);
LOG.info("Copied Job Output to:" + finalPath);
} finally {
// fs.close();
}
} catch (IOException e) {
LOG.error(StringUtils.stringifyException(e));
e.printStackTrace();
}
}
}
