1 | package felix.io; |
2 | import edu.umd.cloud9.collection.XMLInputFormat; |
3 | |
4 | import java.io.IOException; |
5 | import java.util.ArrayList; |
6 | import java.util.Date; |
7 | import java.util.Iterator; |
8 | import java.util.StringTokenizer; |
9 | import org.apache.hadoop.streaming.StreamInputFormat; |
10 | import org.apache.hadoop.streaming.StreamXmlRecordReader; |
11 | |
12 | import org.apache.hadoop.conf.Configuration; |
13 | import org.apache.hadoop.conf.Configured; |
14 | import org.apache.hadoop.fs.FileSystem; |
15 | import org.apache.hadoop.fs.FileUtil; |
16 | import org.apache.hadoop.fs.Path; |
17 | import org.apache.hadoop.io.*; |
18 | import org.apache.hadoop.mapred.JobClient; |
19 | import org.apache.hadoop.mapred.OutputCollector; |
20 | import org.apache.hadoop.mapred.Reporter; |
21 | import org.apache.hadoop.mapred.lib.db.DBInputFormat; |
22 | |
23 | import org.apache.hadoop.mapreduce.Job; |
24 | import org.apache.hadoop.mapreduce.Mapper; |
25 | import org.apache.hadoop.mapreduce.Reducer; |
26 | import org.apache.hadoop.mapreduce.lib.input.FileInputFormat; |
27 | import org.apache.hadoop.mapreduce.lib.input.FileSplit; |
28 | import org.apache.hadoop.mapreduce.lib.input.TextInputFormat; |
29 | import org.apache.hadoop.mapreduce.lib.output.FileOutputFormat; |
30 | import org.apache.hadoop.mapreduce.lib.output.TextOutputFormat; |
31 | |
32 | |
33 | import org.apache.hadoop.util.Tool; |
34 | import org.apache.hadoop.util.ToolRunner; |
35 | import org.python.core.PyArray; |
36 | import org.python.core.PyList; |
37 | import org.python.core.PyLong; |
38 | import org.python.core.PyObject; |
39 | import org.python.core.PyString; |
40 | |
41 | import tuffy.util.ExceptionMan; |
42 | |
43 | import felix.thirdpart.XmlInputFormat; |
44 | import felix.util.FelixConfig; |
45 | |
46 | /** |
47 | * This class communicates with Hadoop. It translates |
48 | * the given MAP and REDUCE codes to Hadoop API. |
49 | * @author czhang |
50 | * |
51 | */ |
52 | public class TestHadoop extends Configured implements Tool{ |
53 | |
54 | /** |
55 | * MAP |
56 | * @author czhang |
57 | * |
58 | */ |
59 | public static class Map extends Mapper<LongWritable, Text, Text, IntWritable>{ |
60 | |
61 | private Text word = new Text(); |
62 | public static PythonExecutor pyMap; |
63 | |
64 | public static String _inputvalue; |
65 | |
66 | @Override |
67 | public void setup(Mapper.Context contex) { |
68 | |
69 | pyMap = new PythonExecutor( |
70 | "_felix_donotusemyname_outkey=[]\n" + "_felix_donotusemyname_outvalue=[]\n"+ |
71 | contex.getConfiguration().get("pyMapScript")); |
72 | _inputvalue = contex.getConfiguration().get("mapinputvalue"); |
73 | |
74 | if(contex.getConfiguration().get("pyMapInitScript") != null){ |
75 | pyMap.execSingle(contex.getConfiguration().get("pyMapInitScript")); |
76 | } |
77 | |
78 | } |
79 | |
80 | public void map(LongWritable key, Text value, Mapper.Context context) throws IOException, InterruptedException { |
81 | |
82 | FileSplit fileSplit = (FileSplit)context.getInputSplit(); |
83 | String fileName = fileSplit.getPath().getName(); |
84 | |
85 | pyMap.set("_offset", new PyLong(key.toString())); |
86 | pyMap.set("_filename", new PyString(fileName.toString())); |
87 | pyMap.set(_inputvalue, new PyString(value.toString())); |
88 | |
89 | pyMap.run(); |
90 | PyList outKey = (PyList) pyMap.get("_felix_donotusemyname_outkey"); |
91 | PyList outValue = (PyList) pyMap.get("_felix_donotusemyname_outvalue"); |
92 | |
93 | for(int i=0;i<outKey.__len__();i++){ |
94 | PyObject _key = outKey.__getitem__(i); |
95 | PyObject _value = outValue.__getitem__(i); |
96 | |
97 | context.write(new Text(""+_key), new Text(""+_value)); |
98 | // System.out.println(""+key + "\t" + value); |
99 | } |
100 | |
101 | } |
102 | |
103 | } |
104 | |
105 | /** |
106 | * REDUCE |
107 | * @author czhang |
108 | * |
109 | */ |
110 | public static class Reduce extends Reducer<Text, Text, Text, Text>{ |
111 | |
112 | public static PythonExecutor pyReducer; |
113 | |
114 | public static String _inputkey; |
115 | public static String _inputvalues; |
116 | |
117 | @Override |
118 | public void setup(Reducer.Context contex) { |
119 | pyReducer = new PythonExecutor( |
120 | "_felix_donotusemyname_outkey=[]\n" + "_felix_donotusemyname_outvalue=[]\n" + |
121 | contex.getConfiguration().get("pyReduceScript")); |
122 | _inputkey = contex.getConfiguration().get("reduceinputkey"); |
123 | _inputvalues = contex.getConfiguration().get("reduceinputvalues"); |
124 | |
125 | if(contex.getConfiguration().get("pyReduceInitScript") != null){ |
126 | pyReducer.execSingle(contex.getConfiguration().get("pyReduceInitScript")); |
127 | } |
128 | |
129 | } |
130 | |
131 | protected void reduce(Text _key, Iterable _values, Context context) throws IOException, InterruptedException{ |
132 | |
133 | |
134 | ArrayList<PyString> args = new ArrayList<PyString>(); |
135 | Iterator it = _values.iterator(); |
136 | while(it.hasNext()){ |
137 | args.add(new PyString(it.next().toString())); |
138 | } |
139 | |
140 | PyList values = new PyList(args); |
141 | PyObject inputkey = new PyString(_key.toString()); |
142 | |
143 | pyReducer.set(_inputvalues, values); |
144 | pyReducer.set(_inputkey, inputkey); |
145 | |
146 | pyReducer.run(); |
147 | PyList outKey = (PyList) pyReducer.get("_felix_donotusemyname_outkey"); |
148 | PyList outValue = (PyList) pyReducer.get("_felix_donotusemyname_outvalue"); |
149 | |
150 | for(int i=0;i<outKey.__len__();i++){ |
151 | PyObject key = outKey.__getitem__(i); |
152 | PyObject value = outValue.__getitem__(i); |
153 | context.write(new Text("True\t1\t"+key), new Text(""+value)); |
154 | } |
155 | |
156 | } |
157 | |
158 | } |
159 | |
160 | /** |
161 | * @deprecated |
162 | * @param mapScript |
163 | * @param reduceScript |
164 | */ |
165 | public TestHadoop(String mapScript, String reduceScript){ |
166 | Map.pyMap = new PythonExecutor( |
167 | mapScript); |
168 | |
169 | Reduce.pyReducer = new PythonExecutor( |
170 | reduceScript); |
171 | } |
172 | |
173 | /** |
174 | * Execute a given task configuration (via arg0). |
175 | * @param toPass |
176 | * @throws Exception |
177 | */ |
178 | public static void executeHadoopProgram(String[] toPass) throws Exception{ |
179 | |
180 | String[] dirtoPass = toPass.clone(); |
181 | dirtoPass[2] += "_dir"; |
182 | int res = ToolRunner.run(new Configuration(), new TestHadoop(dirtoPass[3], dirtoPass[4]), |
183 | dirtoPass); |
184 | |
185 | } |
186 | |
187 | /** |
188 | * @deprecated |
189 | * @param toPass |
190 | * @throws Exception |
191 | */ |
192 | public static void post(String[] toPass) throws Exception{ |
193 | |
194 | Configuration fsconf = new Configuration(); |
195 | fsconf.set("fs.default.name", FelixConfig.hdfsServer); |
196 | FileSystem fileSystem = FileSystem.get(fsconf); |
197 | |
198 | String[] dirtoPass = toPass.clone(); |
199 | dirtoPass[2] += "_dir"; |
200 | |
201 | FileUtil.copyMerge(fileSystem, new Path(dirtoPass[2]), fileSystem, new Path(toPass[2]), true, |
202 | fsconf, ""); |
203 | |
204 | fileSystem.close(); |
205 | |
206 | } |
207 | |
208 | |
209 | /** |
210 | * Test entry. |
211 | * @param args |
212 | * @throws Exception |
213 | */ |
214 | public static void main(String[] args) throws Exception{ |
215 | |
216 | String[] toPass = { |
217 | |
218 | "hdfs://d-02.cs.wisc.edu:9000/felixNE/Entity.db", |
219 | //"hdfs://d-02.cs.wisc.edu:9000/firstMapReduceOut42/part-00000", |
220 | |
221 | "hdfs://d-02.cs.wisc.edu:9000/firstMapReduceOut100", |
222 | |
223 | "for k in _input.split(' '):\n"+ |
224 | "\t_outkey.append(k)\n"+ |
225 | "\t_outvalue.append('1')\n", |
226 | |
227 | "_outkey.append(_inputkey)\n"+ |
228 | "_outvalue.append(len(_inputvalues))\n" |
229 | |
230 | }; |
231 | |
232 | TestHadoop.executeHadoopProgram(toPass); |
233 | |
234 | |
235 | } |
236 | |
237 | |
238 | @Override |
239 | /** |
240 | * Execute a given task configuration (via arg0). |
241 | */ |
242 | public int run(String[] arg0) throws Exception { |
243 | |
244 | |
245 | Configuration conf = getConf(); |
246 | |
247 | if(!FelixConfig.hadoopLocal){ |
248 | conf.set("fs.defaultFS", FelixConfig.hdfsServer); |
249 | conf.set("mapred.job.tracker", FelixConfig.mrServer); |
250 | conf.set("mapred.child.java.opts", "-Xmx3192m"); |
251 | } |
252 | conf.set("pyMapScript", arg0[3]); |
253 | conf.set("pyReduceScript", arg0[4]); |
254 | conf.set("mapinputvalue", arg0[5]); |
255 | conf.set("reduceinputkey", arg0[6]); |
256 | conf.set("reduceinputvalues", arg0[7]); |
257 | conf.set("pyMapInitScript", arg0[8]); |
258 | conf.set("pyReduceInitScript", arg0[9]); |
259 | |
260 | |
261 | if(arg0[0].equals("xml")){ |
262 | conf.set("xmlinput.start", arg0[10]); |
263 | conf.set("xmlinput.end", arg0[11]); |
264 | } |
265 | |
266 | Job job = new Job(conf, "Felix_Run_On" + (new Date()).toLocaleString()); |
267 | |
268 | job.setNumReduceTasks(FelixConfig.nReduce); |
269 | |
270 | job.setJarByClass(TestHadoop.class); |
271 | |
272 | job.setMapperClass(Map.class); |
273 | job.setReducerClass(Reduce.class); |
274 | |
275 | job.setOutputKeyClass(Text.class); |
276 | job.setOutputValueClass(Text.class); |
277 | |
278 | FileInputFormat.addInputPath(job, new Path(arg0[1])); |
279 | FileOutputFormat.setOutputPath(job, new Path(arg0[2])); |
280 | |
281 | |
282 | if(arg0[0].equals("xml")){ |
283 | //XML |
284 | |
285 | job.setInputFormatClass(XmlInputFormat.class); |
286 | job.setOutputFormatClass(TextOutputFormat.class); |
287 | |
288 | job.setOutputKeyClass(Text.class); |
289 | job.setOutputValueClass(Text.class); |
290 | |
291 | }else if(arg0[0].equals("standard")){ |
292 | |
293 | job.setInputFormatClass(TextInputFormat.class); |
294 | job.setOutputFormatClass(TextOutputFormat.class); |
295 | |
296 | job.setOutputKeyClass(Text.class); |
297 | job.setOutputValueClass(Text.class); |
298 | } |
299 | |
300 | job.waitForCompletion(true); |
301 | return 0; |
302 | } |
303 | |
304 | } |