public static int main(String[] args) throws Exception {
int i;
String outPath;
int numMaps = 0, numReds = 0;
List<String> other_args = new ArrayList<String>();
for (i = 0; i < args.length; ++i) {
try {
if ("-m".equals(args[i])) {
numMaps = Integer.parseInt(args[++i]);
} else if ("-r".equals(args[i])) {
numReds = Integer.parseInt(args[++i]);
} else {
other_args.add(args[i]);
}
} catch (NumberFormatException except) {
System.out.println("ERROR: Integer expected instead of " + args[i]);
printUsage();
} catch (ArrayIndexOutOfBoundsException except) {
System.out.println("ERROR: Required parameter missing from " + args[i - 1]);
printUsage(); // exits
}
}
// Make sure there are exactly 2 parameters left.
if (other_args.size() != 2) {
System.out.println(
"ERROR: Wrong number of parameters: " + other_args.size() + " instead of 2.");
printUsage();
}
Date startTime = new Date();
System.out.println("Job started: " + startTime);
Date startIteration;
Date endIteration;
JobConf conf = new JobConf(Kmeans.class);
conf.setJobName("kmeans");
conf.setOutputKeyClass(IntWritable.class);
conf.setOutputValueClass(Text.class);
conf.setMapOutputKeyClass(IntWritable.class);
conf.setMapOutputValueClass(ClusterWritable.class);
conf.setMapperClass(MapClass.class);
conf.setReducerClass(Reduce.class);
conf.setNumMapTasks(numMaps);
conf.setNumReduceTasks(numReds);
FileInputFormat.setInputPaths(conf, new Path(other_args.get(0)));
outPath = new String(other_args.get(1));
FileOutputFormat.setOutputPath(conf, new Path(outPath));
startIteration = new Date();
JobClient.runJob(conf);
endIteration = new Date();
System.out.println(
"The iteration took "
+ (endIteration.getTime() - startIteration.getTime()) / 1000
+ " seconds.");
return 0;
}
public void reduce(
Text key, Iterator<Text> values, OutputCollector<Text, Text> output, Reporter reporter)
throws IOException {
List<String[]> _temp = new ArrayList<String[]>();
int count = 0;
while (values.hasNext()) {
Text _out = values.next();
String[] tokens = StringUtils.splitPreserveAllTokens(_out.toString(), TAB);
_temp.add(tokens);
if (count++ > 100000) break;
}
if (count > 10000) {
Set<String> ipSet = new HashSet<String>();
for (int posI = 0; posI < _temp.size(); posI++) {
String[] array = _temp.get(posI);
if (array == null) continue;
String mid = array[2];
String ip = array[3];
ipSet.add(ip);
}
output.collect(
key, Utils.mergeKey(String.valueOf(ipSet.size()), StringUtils.join(ipSet, '|')));
return;
}
/**
* ·Ö×éËã·¨ FOREACH ALL_DATA IF IN INDEX THEN UPDATE INDEX AND INSERT DATA ELSE FOREACH SUB_DATA
* MAKE INDEX AND SET FIND'S DATA AS NULL
*/
// List<List<String[]>> dataList = new ArrayList<List<String[]>>();
List<StringBuffer> indexList = new ArrayList<StringBuffer>();
Set<String> ipSet = new HashSet<String>();
boolean muliHost = false;
for (int posI = 0; posI < _temp.size(); posI++) {
String[] array = _temp.get(posI);
if (array == null) continue;
String mid = array[2];
String ip = array[3];
ipSet.add(ip);
boolean hasIndex = false;
for (int i = 0; i < indexList.size(); i++) {
StringBuffer index = indexList.get(i);
if (index.indexOf("|" + mid + "|") >= 0 || index.indexOf("|" + ip + "|") >= 0) {
if (index.indexOf("|" + mid + "|") < 0) {
index.append('|').append(mid).append('|');
}
if (index.indexOf("|" + ip + "|") < 0) {
index.append('|').append(ip).append('|');
}
// dataList.get(i).add(array);
hasIndex = true;
break;
}
}
if (!hasIndex) {
StringBuffer index = new StringBuffer("|" + mid + "|" + ip + "|");
// List<String[]> _tmp = new ArrayList<String[]>();
// _tmp.add(array);
for (int k = posI + 1; k < _temp.size(); k++) {
String[] _newArray = _temp.get(k);
if (_newArray == null) {
continue;
}
String _mid = _newArray[2];
String _ip = _newArray[3];
if (index.indexOf("|" + _mid + "|") >= 0 || index.indexOf("|" + _ip + "|") >= 0) {
if (index.indexOf("|" + _mid + "|") < 0) {
index.append('|').append(_mid).append('|');
}
if (index.indexOf("|" + _ip + "|") < 0) {
index.append('|').append(_ip).append('|');
}
// _tmp.add(_newArray);
_temp.set(k, null);
}
}
indexList.add(index);
// dataList.add(_tmp);
}
}
// for(String[] _array : _temp){
// output.collect(key,Utils.mergeKey(_array[1],_array[2],_array[3],_array[4]));
// }
StringBuffer allIndex = new StringBuffer();
for (StringBuffer index : indexList) {
allIndex.append(index).append(';');
}
if (allIndex.length() > 0) {
allIndex.deleteCharAt(allIndex.length() - 1);
}
output.collect(
key, Utils.mergeKey(String.valueOf(indexList.size()), StringUtils.join(ipSet, '|')));
}