/**
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The ASF licenses this file to You under the Apache License, Version 2.0
* (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package org.apache.mahout.text;
import java.io.IOException;
import java.nio.charset.Charset;
import java.util.Map;
import com.google.common.collect.Maps;
import com.google.common.io.Closeables;
import org.apache.commons.lang3.StringUtils;
import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.FileStatus;
import org.apache.hadoop.fs.FileSystem;
import org.apache.hadoop.fs.Path;
import org.apache.hadoop.fs.PathFilter;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapreduce.Job;
import org.apache.hadoop.mapreduce.lib.input.FileInputFormat;
import org.apache.hadoop.mapreduce.lib.output.FileOutputFormat;
import org.apache.hadoop.mapreduce.lib.output.SequenceFileOutputFormat;
import org.apache.hadoop.util.ToolRunner;
import org.apache.mahout.common.AbstractJob;
import org.apache.mahout.common.ClassUtils;
import org.apache.mahout.common.HadoopUtil;
import org.apache.mahout.common.commandline.DefaultOptionCreator;
import org.apache.mahout.utils.io.ChunkedWriter;
/**
* Converts a directory of text documents into SequenceFiles of Specified chunkSize. This class takes in a
* parent directory containing sub folders of text documents and recursively reads the files and creates the
* {@link org.apache.hadoop.io.SequenceFile}s of docid => content. The docid is set as the relative path of the
* document from the parent directory prepended with a specified prefix. You can also specify the input encoding
* of the text files. The content of the output SequenceFiles are encoded as UTF-8 text.
*/
public class SequenceFilesFromDirectory extends AbstractJob {
private static final String PREFIX_ADDITION_FILTER = PrefixAdditionFilter.class.getName();
private static final String[] CHUNK_SIZE_OPTION = {"chunkSize", "chunk"};
public static final String[] FILE_FILTER_CLASS_OPTION = {"fileFilterClass", "filter"};
private static final String[] CHARSET_OPTION = {"charset", "c"};
private static final int MAX_JOB_SPLIT_LOCATIONS = 1000000;
public static final String[] KEY_PREFIX_OPTION = {"keyPrefix", "prefix"};
public static final String BASE_INPUT_PATH = "baseinputpath";
public static void main(String[] args) throws Exception {
ToolRunner.run(new SequenceFilesFromDirectory(), args);
}
/*
* callback main after processing MapReduce parameters
*/
@Override
public int run(String[] args) throws Exception {
addOptions();
addOption(DefaultOptionCreator.methodOption().create());
addOption(DefaultOptionCreator.overwriteOption().create());
if (parseArguments(args) == null) {
return -1;
}
Map<String, String> options = parseOptions();
Path output = getOutputPath();
if (hasOption(DefaultOptionCreator.OVERWRITE_OPTION)) {
HadoopUtil.delete(getConf(), output);
}
if (getOption(DefaultOptionCreator.METHOD_OPTION,
DefaultOptionCreator.MAPREDUCE_METHOD).equals(DefaultOptionCreator.SEQUENTIAL_METHOD)) {
runSequential(getConf(), getInputPath(), output, options);
} else {
runMapReduce(getInputPath(), output);
}
return 0;
}
private int runSequential(Configuration conf, Path input, Path output, Map<String, String> options)
throws IOException, InterruptedException, NoSuchMethodException {
// Running sequentially
Charset charset = Charset.forName(getOption(CHARSET_OPTION[0]));
String keyPrefix = getOption(KEY_PREFIX_OPTION[0]);
FileSystem fs = FileSystem.get(input.toUri(), conf);
ChunkedWriter writer = new ChunkedWriter(conf, Integer.parseInt(options.get(CHUNK_SIZE_OPTION[0])), output);
try {
SequenceFilesFromDirectoryFilter pathFilter;
String fileFilterClassName = options.get(FILE_FILTER_CLASS_OPTION[0]);
if (PrefixAdditionFilter.class.getName().equals(fileFilterClassName)) {
pathFilter = new PrefixAdditionFilter(conf, keyPrefix, options, writer, charset, fs);
} else {
pathFilter = ClassUtils.instantiateAs(fileFilterClassName, SequenceFilesFromDirectoryFilter.class,
new Class[] {Configuration.class, String.class, Map.class, ChunkedWriter.class, Charset.class, FileSystem.class},
new Object[] {conf, keyPrefix, options, writer, charset, fs});
}
fs.listStatus(input, pathFilter);
} finally {
Closeables.close(writer, false);
}
return 0;
}
private int runMapReduce(Path input, Path output) throws IOException, ClassNotFoundException, InterruptedException {
int chunkSizeInMB = 64;
if (hasOption(CHUNK_SIZE_OPTION[0])) {
chunkSizeInMB = Integer.parseInt(getOption(CHUNK_SIZE_OPTION[0]));
}
String keyPrefix = null;
if (hasOption(KEY_PREFIX_OPTION[0])) {
keyPrefix = getOption(KEY_PREFIX_OPTION[0]);
}
String fileFilterClassName = null;
if (hasOption(FILE_FILTER_CLASS_OPTION[0])) {
fileFilterClassName = getOption(FILE_FILTER_CLASS_OPTION[0]);
}
PathFilter pathFilter = null;
// Prefix Addition is presently handled in the Mapper and unlike runsequential()
// need not be done via a pathFilter
if (!StringUtils.isBlank(fileFilterClassName) && !PrefixAdditionFilter.class.getName().equals(fileFilterClassName)) {
try {
pathFilter = (PathFilter) Class.forName(fileFilterClassName).newInstance();
} catch (InstantiationException e) {
throw new IllegalStateException(e);
} catch (IllegalAccessException e) {
throw new IllegalStateException(e);
}
}
// Prepare Job for submission.
Job job = prepareJob(input, output, MultipleTextFileInputFormat.class,
SequenceFilesFromDirectoryMapper.class, Text.class, Text.class,
SequenceFileOutputFormat.class, "SequenceFilesFromDirectory");
Configuration jobConfig = job.getConfiguration();
jobConfig.set(KEY_PREFIX_OPTION[0], keyPrefix);
jobConfig.set(FILE_FILTER_CLASS_OPTION[0], fileFilterClassName);
FileSystem fs = FileSystem.get(jobConfig);
FileStatus fsFileStatus = fs.getFileStatus(input);
String inputDirList;
if (pathFilter != null) {
inputDirList = HadoopUtil.buildDirList(fs, fsFileStatus, pathFilter);
} else {
inputDirList = HadoopUtil.buildDirList(fs, fsFileStatus);
}
jobConfig.set(BASE_INPUT_PATH, input.toString());
long chunkSizeInBytes = chunkSizeInMB * 1024 * 1024;
// set the max split locations, otherwise we get nasty debug stuff
jobConfig.set("mapreduce.job.max.split.locations", String.valueOf(MAX_JOB_SPLIT_LOCATIONS));
FileInputFormat.setInputPaths(job, inputDirList);
// need to set this to a multiple of the block size, or no split happens
FileInputFormat.setMaxInputSplitSize(job, chunkSizeInBytes);
FileOutputFormat.setCompressOutput(job, true);
boolean succeeded = job.waitForCompletion(true);
if (!succeeded) {
return -1;
}
return 0;
}
/**
* Override this method in order to add additional options to the command line of the SequenceFileFromDirectory job.
* Do not forget to call super() otherwise all standard options (input/output dirs etc) will not be available.
*/
protected void addOptions() {
addInputOption();
addOutputOption();
addOption(DefaultOptionCreator.overwriteOption().create());
addOption(DefaultOptionCreator.methodOption().create());
addOption(CHUNK_SIZE_OPTION[0], CHUNK_SIZE_OPTION[1], "The chunkSize in MegaBytes. Defaults to 64", "64");
addOption(FILE_FILTER_CLASS_OPTION[0], FILE_FILTER_CLASS_OPTION[1],
"The name of the class to use for file parsing. Default: " + PREFIX_ADDITION_FILTER, PREFIX_ADDITION_FILTER);
addOption(KEY_PREFIX_OPTION[0], KEY_PREFIX_OPTION[1], "The prefix to be prepended to the key", "");
addOption(CHARSET_OPTION[0], CHARSET_OPTION[1],
"The name of the character encoding of the input files. Default to UTF-8", "UTF-8");
}
/**
* Override this method in order to parse your additional options from the command line. Do not forget to call
* super() otherwise standard options (input/output dirs etc) will not be available.
*
* @return Map of options
*/
protected Map<String, String> parseOptions() {
Map<String, String> options = Maps.newHashMap();
options.put(CHUNK_SIZE_OPTION[0], getOption(CHUNK_SIZE_OPTION[0]));
options.put(FILE_FILTER_CLASS_OPTION[0], getOption(FILE_FILTER_CLASS_OPTION[0]));
options.put(CHARSET_OPTION[0], getOption(CHARSET_OPTION[0]));
return options;
}
}