use of org.apache.commons.cli.CommandLine in project Cloud9 by lintool.
the class WikipediaPagesBz2InputStream method main.
@SuppressWarnings("static-access")
public static void main(String[] args) throws Exception {
Options options = new Options();
options.addOption(OptionBuilder.withArgName("path").hasArg().withDescription("gzipped XML dump file").create(INPUT_OPTION));
options.addOption(OptionBuilder.withArgName("lang").hasArg().withDescription("output location").create(LANGUAGE_OPTION));
CommandLine cmdline = null;
CommandLineParser parser = new GnuParser();
try {
cmdline = parser.parse(options, args);
} catch (ParseException exp) {
System.err.println("Error parsing command line: " + exp.getMessage());
System.exit(-1);
}
if (!cmdline.hasOption(INPUT_OPTION)) {
HelpFormatter formatter = new HelpFormatter();
formatter.printHelp(WikipediaPagesBz2InputStream.class.getCanonicalName(), options);
ToolRunner.printGenericCommandUsage(System.out);
System.exit(-1);
}
String path = cmdline.getOptionValue(INPUT_OPTION);
String lang = cmdline.hasOption(LANGUAGE_OPTION) ? cmdline.getOptionValue(LANGUAGE_OPTION) : "en";
WikipediaPage p = WikipediaPageFactory.createWikipediaPage(lang);
WikipediaPagesBz2InputStream stream = new WikipediaPagesBz2InputStream(path);
while (stream.readNext(p)) {
System.out.println(p.getTitle() + "\t" + p.getDocid());
}
}
use of org.apache.commons.cli.CommandLine in project Cloud9 by lintool.
the class DumpWikipediaToPlainText method run.
@SuppressWarnings("static-access")
@Override
public int run(String[] args) throws Exception {
Options options = new Options();
options.addOption(OptionBuilder.withArgName("path").hasArg().withDescription("XML dump file").create(INPUT_OPTION));
options.addOption(OptionBuilder.withArgName("path").hasArg().withDescription("output path").create(OUTPUT_OPTION));
options.addOption(OptionBuilder.withArgName("en|sv|de|cs|es|zh|ar|tr").hasArg().withDescription("two-letter language code").create(LANGUAGE_OPTION));
CommandLine cmdline;
CommandLineParser parser = new GnuParser();
try {
cmdline = parser.parse(options, args);
} catch (ParseException exp) {
System.err.println("Error parsing command line: " + exp.getMessage());
return -1;
}
if (!cmdline.hasOption(INPUT_OPTION) || !cmdline.hasOption(OUTPUT_OPTION)) {
HelpFormatter formatter = new HelpFormatter();
formatter.printHelp(this.getClass().getName(), options);
ToolRunner.printGenericCommandUsage(System.out);
return -1;
}
// Assume "en" by default.
String language = "en";
if (cmdline.hasOption(LANGUAGE_OPTION)) {
language = cmdline.getOptionValue(LANGUAGE_OPTION);
if (language.length() != 2) {
System.err.println("Error: \"" + language + "\" unknown language!");
return -1;
}
}
String inputPath = cmdline.getOptionValue(INPUT_OPTION);
String outputPath = cmdline.getOptionValue(OUTPUT_OPTION);
LOG.info("Tool name: " + this.getClass().getName());
LOG.info(" - XML dump file: " + inputPath);
LOG.info(" - output path: " + outputPath);
LOG.info(" - language: " + language);
Job job = Job.getInstance(getConf());
job.setJarByClass(DumpWikipediaToPlainText.class);
job.setJobName(String.format("DumpWikipediaToPlainText[%s: %s, %s: %s, %s: %s]", INPUT_OPTION, inputPath, OUTPUT_OPTION, outputPath, LANGUAGE_OPTION, language));
job.setNumReduceTasks(0);
FileInputFormat.setInputPaths(job, new Path(inputPath));
FileOutputFormat.setOutputPath(job, new Path(outputPath));
if (language != null) {
job.getConfiguration().set("wiki.language", language);
}
job.setInputFormatClass(WikipediaPageInputFormat.class);
job.setOutputFormatClass(TextOutputFormat.class);
job.setMapperClass(MyMapper.class);
job.setOutputKeyClass(Text.class);
job.setOutputValueClass(Text.class);
// Delete the output directory if it exists already.
FileSystem.get(getConf()).delete(new Path(outputPath), true);
job.waitForCompletion(true);
return 0;
}
use of org.apache.commons.cli.CommandLine in project Cloud9 by lintool.
the class WikipediaDocnoMappingBuilder method run.
@SuppressWarnings("static-access")
@Override
public int run(String[] args) throws Exception {
Options options = new Options();
options.addOption(OptionBuilder.withArgName("path").hasArg().withDescription("XML dump file").create(INPUT_OPTION));
options.addOption(OptionBuilder.withArgName("path").hasArg().withDescription("output file").create(OUTPUT_FILE_OPTION));
options.addOption(OptionBuilder.withArgName("en|sv|de|cs|es|zh|ar|tr").hasArg().withDescription("two-letter language code").create(LANGUAGE_OPTION));
options.addOption(KEEP_ALL_OPTION, false, "keep all pages");
CommandLine cmdline;
CommandLineParser parser = new GnuParser();
try {
cmdline = parser.parse(options, args);
} catch (ParseException exp) {
System.err.println("Error parsing command line: " + exp.getMessage());
return -1;
}
if (!cmdline.hasOption(INPUT_OPTION) || !cmdline.hasOption(OUTPUT_FILE_OPTION)) {
HelpFormatter formatter = new HelpFormatter();
formatter.printHelp(this.getClass().getName(), options);
ToolRunner.printGenericCommandUsage(System.out);
return -1;
}
String language = null;
if (cmdline.hasOption(LANGUAGE_OPTION)) {
language = cmdline.getOptionValue(LANGUAGE_OPTION);
if (language.length() != 2) {
System.err.println("Error: \"" + language + "\" unknown language!");
return -1;
}
}
String inputPath = cmdline.getOptionValue(INPUT_OPTION);
String outputFile = cmdline.getOptionValue(OUTPUT_FILE_OPTION);
boolean keepAll = cmdline.hasOption(KEEP_ALL_OPTION);
String tmpPath = "tmp-" + WikipediaDocnoMappingBuilder.class.getSimpleName() + "-" + RANDOM.nextInt(10000);
LOG.info("Tool name: " + this.getClass().getName());
LOG.info(" - input: " + inputPath);
LOG.info(" - output file: " + outputFile);
LOG.info(" - keep all pages: " + keepAll);
LOG.info(" - language: " + language);
Job job = Job.getInstance(getConf());
job.setJarByClass(WikipediaDocnoMappingBuilder.class);
job.setJobName(String.format("BuildWikipediaDocnoMapping[%s: %s, %s: %s, %s: %s]", INPUT_OPTION, inputPath, OUTPUT_FILE_OPTION, outputFile, LANGUAGE_OPTION, language));
job.getConfiguration().setBoolean(KEEP_ALL_OPTION, keepAll);
if (language != null) {
job.getConfiguration().set("wiki.language", language);
}
job.setNumReduceTasks(1);
FileInputFormat.setInputPaths(job, new Path(inputPath));
FileOutputFormat.setOutputPath(job, new Path(tmpPath));
FileOutputFormat.setCompressOutput(job, false);
job.setOutputKeyClass(IntWritable.class);
job.setOutputValueClass(IntWritable.class);
job.setInputFormatClass(WikipediaPageInputFormat.class);
job.setOutputFormatClass(TextOutputFormat.class);
job.setMapperClass(MyMapper.class);
job.setReducerClass(MyReducer.class);
// Delete the output directory if it exists already.
FileSystem.get(getConf()).delete(new Path(tmpPath), true);
job.waitForCompletion(true);
long cnt = keepAll ? job.getCounters().findCounter(PageTypes.TOTAL).getValue() : job.getCounters().findCounter(PageTypes.ARTICLE).getValue();
WikipediaDocnoMapping.writeDocnoMappingData(FileSystem.get(getConf()), tmpPath + "/part-r-00000", (int) cnt, outputFile);
FileSystem.get(getConf()).delete(new Path(tmpPath), true);
return 0;
}
use of org.apache.commons.cli.CommandLine in project Cloud9 by lintool.
the class TrecForwardIndexBuilder method run.
/**
* Runs this tool.
*/
@SuppressWarnings("static-access")
public int run(String[] args) throws Exception {
Options options = new Options();
options.addOption(OptionBuilder.withArgName("path").hasArg().withDescription("(required) collection path").create(COLLECTION_OPTION));
options.addOption(OptionBuilder.withArgName("path").hasArg().withDescription("(required) output index path").create(INDEX_OPTION));
options.addOption(OptionBuilder.withArgName("path").hasArg().withDescription("(required) DocnoMapping data").create(MAPPING_OPTION));
CommandLine cmdline;
CommandLineParser parser = new GnuParser();
try {
cmdline = parser.parse(options, args);
} catch (ParseException exp) {
System.err.println("Error parsing command line: " + exp.getMessage());
return -1;
}
if (!cmdline.hasOption(COLLECTION_OPTION) || !cmdline.hasOption(INDEX_OPTION) || !cmdline.hasOption(MAPPING_OPTION)) {
HelpFormatter formatter = new HelpFormatter();
formatter.printHelp(this.getClass().getName(), options);
ToolRunner.printGenericCommandUsage(System.out);
return -1;
}
String collectionPath = cmdline.getOptionValue(COLLECTION_OPTION);
String indexFile = cmdline.getOptionValue(INDEX_OPTION);
String mappingFile = cmdline.getOptionValue(MAPPING_OPTION);
String tmpDir = "tmp-" + TrecForwardIndexBuilder.class.getSimpleName() + "-" + random.nextInt(10000);
Job job = new Job(getConf(), TrecForwardIndexBuilder.class.getSimpleName() + ":" + collectionPath);
job.setJarByClass(TrecForwardIndexBuilder.class);
FileSystem fs = FileSystem.get(getConf());
LOG.info("Tool name: " + TrecForwardIndexBuilder.class.getSimpleName());
LOG.info(" - collection path: " + collectionPath);
LOG.info(" - index file: " + indexFile);
LOG.info(" - DocnoMapping file: " + mappingFile);
LOG.info(" - temp output directory: " + tmpDir);
job.setNumReduceTasks(1);
if (job.getConfiguration().get("mapred.job.tracker").equals("local")) {
job.getConfiguration().set(DOCNO_MAPPING_FILE_PROPERTY, mappingFile);
} else {
DistributedCache.addCacheFile(new URI(mappingFile), job.getConfiguration());
}
FileInputFormat.setInputPaths(job, new Path(collectionPath));
FileOutputFormat.setOutputPath(job, new Path(tmpDir));
FileOutputFormat.setCompressOutput(job, false);
job.setInputFormatClass(TrecDocumentInputFormat.class);
job.setOutputKeyClass(IntWritable.class);
job.setOutputValueClass(Text.class);
job.setMapperClass(MyMapper.class);
// delete the output directory if it exists already
FileSystem.get(getConf()).delete(new Path(tmpDir), true);
job.waitForCompletion(true);
Counters counters = job.getCounters();
int numDocs = (int) counters.findCounter(Count.DOCS).getValue();
String inputFile = tmpDir + "/" + "part-r-00000";
LOG.info("Writing " + numDocs + " doc offseta to " + indexFile);
LineReader reader = new LineReader(fs.open(new Path(inputFile)));
FSDataOutputStream writer = fs.create(new Path(indexFile), true);
writer.writeUTF(edu.umd.cloud9.collection.trec.TrecForwardIndex.class.getCanonicalName());
writer.writeUTF(collectionPath);
writer.writeInt(numDocs);
int cnt = 0;
Text line = new Text();
while (reader.readLine(line) > 0) {
String[] arr = line.toString().split("\\t");
long offset = Long.parseLong(arr[1]);
int len = Integer.parseInt(arr[2]);
writer.writeLong(offset);
writer.writeInt(len);
cnt++;
if (cnt % 100000 == 0) {
LOG.info(cnt + " docs");
}
}
reader.close();
writer.close();
LOG.info(cnt + " docs total. Done!");
if (numDocs != cnt) {
throw new RuntimeException("Unexpected number of documents in building forward index!");
}
fs.delete(new Path(tmpDir), true);
return 0;
}
use of org.apache.commons.cli.CommandLine in project Cloud9 by lintool.
the class CountMedlineCitations method run.
@SuppressWarnings("static-access")
public int run(String[] args) throws Exception {
Options options = new Options();
options.addOption(OptionBuilder.withArgName("path").hasArg().withDescription("(required) collection path").create(COLLECTION_OPTION));
options.addOption(OptionBuilder.withArgName("path").hasArg().withDescription("(required) output path").create(OUTPUT_OPTION));
options.addOption(OptionBuilder.withArgName("path").hasArg().withDescription("(required) DocnoMapping data").create(MAPPING_OPTION));
CommandLine cmdline;
CommandLineParser parser = new GnuParser();
try {
cmdline = parser.parse(options, args);
} catch (ParseException exp) {
System.err.println("Error parsing command line: " + exp.getMessage());
return -1;
}
if (!cmdline.hasOption(COLLECTION_OPTION) || !cmdline.hasOption(OUTPUT_OPTION) || !cmdline.hasOption(MAPPING_OPTION)) {
HelpFormatter formatter = new HelpFormatter();
formatter.printHelp(this.getClass().getName(), options);
ToolRunner.printGenericCommandUsage(System.out);
return -1;
}
String inputPath = cmdline.getOptionValue(COLLECTION_OPTION);
String outputPath = cmdline.getOptionValue(OUTPUT_OPTION);
String mappingFile = cmdline.getOptionValue(MAPPING_OPTION);
LOG.info("Tool: " + CountMedlineCitations.class.getSimpleName());
LOG.info(" - input: " + inputPath);
LOG.info(" - output dir: " + outputPath);
LOG.info(" - docno mapping file: " + mappingFile);
Job job = new Job(getConf(), CountMedlineCitations.class.getSimpleName() + ":" + inputPath);
job.setJarByClass(CountMedlineCitations.class);
job.setNumReduceTasks(0);
// Pass in the class name as a String; this is makes the mapper general in being able to load
// any collection of Indexable objects that has docid/docno mapping specified by a DocnoMapping
// object.
job.getConfiguration().set("DocnoMappingClass", MedlineDocnoMapping.class.getCanonicalName());
// Put the mapping file in the distributed cache so each map worker will have it.
DistributedCache.addCacheFile(new URI(mappingFile), job.getConfiguration());
FileInputFormat.setInputPaths(job, new Path(inputPath));
FileOutputFormat.setOutputPath(job, new Path(outputPath));
FileOutputFormat.setCompressOutput(job, false);
job.setInputFormatClass(MedlineCitationInputFormat.class);
job.setOutputKeyClass(Text.class);
job.setOutputValueClass(IntWritable.class);
job.setMapperClass(MyMapper.class);
// Delete the output directory if it exists already.
FileSystem.get(job.getConfiguration()).delete(new Path(outputPath), true);
job.waitForCompletion(true);
Counters counters = job.getCounters();
int numDocs = (int) counters.findCounter(Count.DOCS).getValue();
LOG.info("Read " + numDocs + " docs.");
return numDocs;
}
Aggregations