(转)Pig 重写加载函数和存储函数UDF

pig自带的pigstorage不能指定行分隔符，所以自己重写了一个简单的UDF类，可以指定列和行的分隔符，之前研究过的简单的，

http://blog.csdn.net/ruishenh/article/details/12048067

但是弊端大，所以这次重写一下。

 

操作步骤打好包上传到服务器，

grunt> register  /home/pig/pig-0.11.0/udflib/myStorage.jar 

grunt> cat student; 1,xiaohouzi,25/2,xiaohouzi2,24/3,xiaohouzi3,23    

grunt> a = load 'student' using com.hcr.hadoop.pig.MyStorage(',','/');

grunt> dump a;

(1,xiaohouzi,25) (2,xiaohouzi2,24) (3,xiaohouzi3,23)

grunt> store a into 'myStorageOut' using com.hcr.hadoop.pig.MyStorage(',','/');

执行提示成功后查看

grunt> cat myStorageOut 1,xiaohouzi,25/2,xiaohouzi2,24/3,xiaohouzi3,23/

 

 

源码类

  1 package com.hcr.hadoop.pig;
  2 
  3 import java.io.DataOutputStream;
  4 import java.io.IOException;
  5 import java.io.UnsupportedEncodingException;
  6 import java.util.ArrayList;
  7 import java.util.List;
  8 
  9 import org.apache.commons.logging.Log;
 10 import org.apache.commons.logging.LogFactory;
 11 import org.apache.hadoop.conf.Configuration;
 12 import org.apache.hadoop.fs.FSDataOutputStream;
 13 import org.apache.hadoop.fs.FileSystem;
 14 import org.apache.hadoop.fs.Path;
 15 import org.apache.hadoop.io.LongWritable;
 16 import org.apache.hadoop.io.Text;
 17 import org.apache.hadoop.io.WritableComparable;
 18 import org.apache.hadoop.io.compress.BZip2Codec;
 19 import org.apache.hadoop.io.compress.CompressionCodec;
 20 import org.apache.hadoop.io.compress.GzipCodec;
 21 import org.apache.hadoop.mapreduce.InputFormat;
 22 import org.apache.hadoop.mapreduce.InputSplit;
 23 import org.apache.hadoop.mapreduce.Job;
 24 import org.apache.hadoop.mapreduce.OutputFormat;
 25 import org.apache.hadoop.mapreduce.RecordReader;
 26 import org.apache.hadoop.mapreduce.RecordWriter;
 27 import org.apache.hadoop.mapreduce.TaskAttemptContext;
 28 import org.apache.hadoop.mapreduce.lib.input.FileInputFormat;
 29 import org.apache.hadoop.mapreduce.lib.input.LineRecordReader;
 30 import org.apache.hadoop.mapreduce.lib.input.TextInputFormat;
 31 import org.apache.hadoop.mapreduce.lib.output.FileOutputFormat;
 32 import org.apache.hadoop.mapreduce.lib.output.TextOutputFormat;
 33 import org.apache.hadoop.util.ReflectionUtils;
 34 import org.apache.pig.Expression;
 35 import org.apache.pig.LoadFunc;
 36 import org.apache.pig.LoadMetadata;
 37 import org.apache.pig.PigException;
 38 import org.apache.pig.ResourceSchema;
 39 import org.apache.pig.ResourceSchema.ResourceFieldSchema;
 40 import org.apache.pig.ResourceStatistics;
 41 import org.apache.pig.StoreFunc;
 42 import org.apache.pig.StoreFuncInterface;
 43 import org.apache.pig.backend.executionengine.ExecException;
 44 import org.apache.pig.backend.hadoop.executionengine.mapReduceLayer.PigSplit;
 45 import org.apache.pig.data.DataType;
 46 import org.apache.pig.data.Tuple;
 47 import org.apache.pig.data.TupleFactory;
 48 import org.apache.pig.impl.logicalLayer.schema.Schema.FieldSchema;
 49 import org.apache.pig.impl.util.StorageUtil;
 50 
 51 public class MyStorage extends LoadFunc implements StoreFuncInterface,LoadMetadata {
 52 
 53     private static final Log LOG = LogFactory.getLog(MyStorage.class);
 54 
 55     private static final String utf8 = "UTF-8";
 56 
 57     private static String fieldDel = "\t";
 58 
 59     private static String recordDel = "\n";
 60 
 61     protected RecordReader recordReader = null;
 62 
 63     protected RecordWriter writer = null;
 64 
 65     public MyStorage() {
 66     }
 67 
 68     public MyStorage(String fieldDel) {
 69         this(fieldDel, "\n");
 70     }
 71 
 72     public MyStorage(String fieldDel, String recordDel) {
 73         this.fieldDel = fieldDel;
 74         this.recordDel = recordDel;
 75     }
 76 
 77     @Override
 78     public void setLocation(String s, Job job) throws IOException {
 79         FileInputFormat.setInputPaths(job, s);
 80     }
 81 
 82     @Override
 83     public InputFormat getInputFormat() throws IOException {
 84         return new MyStorageInputFormat(recordDel);
 85     }
 86 
 87     @Override
 88     public void prepareToRead(RecordReader recordReader, PigSplit pigSplit)
 89             throws IOException {
 90         this.recordReader = recordReader;
 91     }
 92 
 93     @Override
 94     public Tuple getNext() throws IOException {
 95         try {
 96             boolean flag = recordReader.nextKeyValue();
 97             if (!flag) {
 98                 return null;
 99             }
100             Text value = (Text) recordReader.getCurrentValue();
101             String[] strArray = value.toString().split(fieldDel);
102             List lst = new ArrayList<String>();
103             int i = 0;
104             for (String singleItem : strArray) {
105                 lst.add(i++, singleItem);
106             }
107             return TupleFactory.getInstance().newTuple(lst);
108         } catch (InterruptedException e) {
109             throw new ExecException("Read data error",
110                     PigException.REMOTE_ENVIRONMENT, e);
111         }
112     }
113 
114     /**
115      * */
116     @Override
117     public String relToAbsPathForStoreLocation(String location, Path curDir)
118             throws IOException {
119         return LoadFunc.getAbsolutePath(location, curDir);
120     }
121 
122     @Override
123     public OutputFormat getOutputFormat() throws IOException {
124         return new MyStorageOutputFormat(StorageUtil.parseFieldDel(fieldDel),
125                 this.recordDel);
126     }
127 
128     @Override
129     public void setStoreLocation(String location, Job job) throws IOException {
130         job.getConfiguration().set("mapred.textoutputformat.separator", "");
131         FileOutputFormat.setOutputPath(job, new Path(location));
132         if ("true".equals(job.getConfiguration().get(
133                 "output.compression.enabled"))) {
134             FileOutputFormat.setCompressOutput(job, true);
135             String codec = job.getConfiguration().get(
136                     "output.compression.codec");
137             try {
138                 FileOutputFormat.setOutputCompressorClass(job,
139                         (Class<? extends CompressionCodec>) Class
140                                 .forName(codec));
141             } catch (ClassNotFoundException e) {
142                 throw new RuntimeException("Class not found: " + codec);
143             }
144         } else {
145             // This makes it so that storing to a directory ending with ".gz" or
146             // ".bz2" works.
147             setCompression(new Path(location), job);
148         }
149 
150     }
151 
152     private void setCompression(Path path, Job job) {
153         String location = path.getName();
154         if (location.endsWith(".bz2") || location.endsWith(".bz")) {
155             FileOutputFormat.setCompressOutput(job, true);
156             FileOutputFormat.setOutputCompressorClass(job, BZip2Codec.class);
157         } else if (location.endsWith(".gz")) {
158             FileOutputFormat.setCompressOutput(job, true);
159             FileOutputFormat.setOutputCompressorClass(job, GzipCodec.class);
160         } else {
161             FileOutputFormat.setCompressOutput(job, false);
162         }
163     }
164 
165     @Override
166     public void checkSchema(ResourceSchema s) throws IOException {
167         // TODO Auto-generated method stub
168 
169     }
170 
171     @Override
172     public void prepareToWrite(RecordWriter writer) throws IOException {
173         this.writer = writer;
174     }
175 
176     @Override
177     public void putNext(Tuple t) throws IOException {
178         try {
179             writer.write(null, t);
180         } catch (InterruptedException e) {
181             throw new IOException(e);
182         }
183     }
184 
185     @Override
186     public void setStoreFuncUDFContextSignature(String signature) {
187         // TODO Auto-generated method stub
188 
189     }
190 
191     @Override
192     public void cleanupOnFailure(String location, Job job) throws IOException {
193         StoreFunc.cleanupOnFailureImpl(location, job);
194     }
195 
196     @Override
197     public void cleanupOnSuccess(String location, Job job) throws IOException {
198         // TODO Auto-generated method stub
199 
200     }
201 
202     @Override
203     public ResourceSchema getSchema(String location, Job job)
204             throws IOException {
205             ResourceSchema rs=new ResourceSchema();
206             FieldSchema c1 = new FieldSchema("c1", DataType.INTEGER);  
207             FieldSchema c2 = new FieldSchema("c2", DataType.INTEGER);
208             FieldSchema c3 = new FieldSchema("c3", DataType.DOUBLE);
209             ResourceFieldSchema fs1 =new ResourceFieldSchema(c1);
210             ResourceFieldSchema fs2 =new ResourceFieldSchema(c2);
211             ResourceFieldSchema fs3 =new ResourceFieldSchema(c3);
212             rs.setFields(new ResourceFieldSchema[]{fs1,fs2,fs3});
213             return rs;
214     }
215 
216     @Override
217     public ResourceStatistics getStatistics(String location, Job job)
218             throws IOException {
219         // TODO Auto-generated method stub
220         return null;
221     }
222 
223     @Override
224     public String[] getPartitionKeys(String location, Job job)
225             throws IOException {
226         // TODO Auto-generated method stub
227         return null;
228     }
229 
230     @Override
231     public void setPartitionFilter(Expression partitionFilter)
232             throws IOException {
233         // TODO Auto-generated method stub
234         
235     }
236 }
237 
238 class MyStorageInputFormat extends TextInputFormat {
239 
240     private final String recordDel;
241 
242     public MyStorageInputFormat(String recordDel) {
243         this.recordDel = recordDel;
244     }
245 
246     @Override
247     public RecordReader<LongWritable, Text> createRecordReader(
248             InputSplit split, TaskAttemptContext context) {
249         String delimiter = context.getConfiguration().get(
250                 "textinputformat.record.delimiter");
251         if (recordDel != null) {
252             delimiter = recordDel;
253         }
254         byte[] recordDelimiterBytes = null;
255         if (null != delimiter){
256             try {
257                 recordDelimiterBytes = decode(delimiter).getBytes("UTF-8");
258             } catch (UnsupportedEncodingException e) {
259                 // TODO Auto-generated catch block
260                 e.printStackTrace();
261             } catch (IOException e) {
262                 // TODO Auto-generated catch block
263                 e.printStackTrace();
264             }
265         }
266         return new LineRecordReader(recordDelimiterBytes);
267     }
268     /**
269      * 工作流传过来的列分隔符,有可能是特殊字符,用八进制或者十六进制表示
270      * @throws IOException 
271      */
272     public static String decode(String str) throws IOException {
273         String re = str;
274         if (str != null && str.startsWith("\\")) {
275             str = str.substring(1, str.length());
276             String[] chars = str.split("\\\\");
277             byte[] bytes = new byte[chars.length];
278             for (int i = 0; i < chars.length; i++) {
279                 if (chars[i].equals("t")) {
280                     bytes[i] = 9;
281                 } else if (chars[i].equals("r")) {
282                     bytes[i] = 13;
283                 } else if (chars[i].equals("n")) {
284                     bytes[i] = 10;
285                 } else if (chars[i].equals("b")) {
286                     bytes[i] = 8;
287                 } else {
288                     bytes[i] = Byte.decode(chars[i]);
289                 }
290             }
291             try {
292                 re = new String(bytes, "UTF-8");
293             } catch (UnsupportedEncodingException e) {
294                 throw new IOException(str, e);
295             }
296         }
297         return re;
298     }
299 
300 }
301 
302 class MyStorageOutputFormat extends TextOutputFormat<WritableComparable, Tuple> {
303 
304     private final byte fieldDel;
305 
306     private final String recordDel;
307 
308     public MyStorageOutputFormat(byte delimiter) {
309         this(delimiter, "\n");
310     }
311 
312     public MyStorageOutputFormat(byte delimiter, String recordDel) {
313         this.fieldDel = delimiter;
314         this.recordDel = recordDel;
315     }
316 
317     protected static class MyRecordWriter extends
318             TextOutputFormat.LineRecordWriter<WritableComparable, Tuple> {
319 
320         private static byte[] newline;
321 
322         private final byte fieldDel;
323 
324         public MyRecordWriter(DataOutputStream out, byte fieldDel)
325                 throws UnsupportedEncodingException {
326             this(out, fieldDel, "\n".getBytes("UTF-8"));
327         }
328 
329         public MyRecordWriter(DataOutputStream out, byte fieldDel, byte[] record) {
330             super(out);
331             this.fieldDel = fieldDel;
332             this.newline = record;
333         }
334 
335         public synchronized void write(WritableComparable key, Tuple value)
336                 throws IOException {
337             int sz = value.size();
338             for (int i = 0; i < sz; i++) {
339                 StorageUtil.putField(out, value.get(i));
340                 if (i != sz - 1) {
341                     out.writeByte(fieldDel);
342                 }
343             }
344             out.write(newline);
345         }
346     }
347 
348     @Override
349     public RecordWriter<WritableComparable, Tuple> getRecordWriter(
350             TaskAttemptContext job) throws IOException, InterruptedException {
351         Configuration conf = job.getConfiguration();
352         boolean isCompressed = getCompressOutput(job);
353         CompressionCodec codec = null;
354         String extension = "";
355         if (isCompressed) {
356             Class<? extends CompressionCodec> codecClass = getOutputCompressorClass(
357                     job, GzipCodec.class);
358             codec = (CompressionCodec) ReflectionUtils.newInstance(codecClass,
359                     conf);
360             extension = codec.getDefaultExtension();
361         }
362         Path file = getDefaultWorkFile(job, extension);
363         FileSystem fs = file.getFileSystem(conf);
364         if (!isCompressed) {
365             FSDataOutputStream fileOut = fs.create(file, false);
366             return new MyRecordWriter(fileOut, fieldDel,
367                     this.recordDel.getBytes());
368         } else {
369             FSDataOutputStream fileOut = fs.create(file, false);
370             return new MyRecordWriter(new DataOutputStream(
371                     codec.createOutputStream(fileOut)), fieldDel,
372                     this.recordDel.getBytes());
373         }
374     }
375 
376 }

grunt> register  /home/pig/pig-0.11.0/udflib/myStorage.jar 
grunt> cat X;
keyDataKNZKCZY:ZDKJS:616150:AFS:3842708d_20131219194420-642464756keyDataKNZKCZY:ZDKJS:616614:AFS:3843920d_20131219194420-642464756keyDataKNZKCZY:ZDKJS:616661:AFS:3844040d_20131219194420-642464756
grunt> a = load 'X' using com.hcr.hadoop.pig.MyStorage('\\001','\\002');
grunt> dump a;
(keyData,KNZKCZY:ZDKJS:616150:AFS:3842708,d_20131219194420-642464756)
(keyData,KNZKCZY:ZDKJS:616614:AFS:3843920,d_20131219194420-642464756)
(keyData,KNZKCZY:ZDKJS:616661:AFS:3844040,d_20131219194420-642464756)
grunt> 

有的时候如果加载模式不想指定具体模式（比如太多了字段，或者不够公有化）就想使用已存在的模式

实现LoadMetadata接口，然后

重写

 

@Override

public ResourceSchema getSchema(String location, Job job)throws IOException {

ResourceSchema rs=new ResourceSchema();            

FieldSchema c1 = new FieldSchema("c1", DataType.INTEGER);              

FieldSchema c2 = new FieldSchema("c2", DataType.INTEGER);            

FieldSchema c3 = new FieldSchema("c3", DataType.DOUBLE);            

ResourceFieldSchema fs1 =new ResourceFieldSchema(c1);            

ResourceFieldSchema fs2 =new ResourceFieldSchema(c2);            

ResourceFieldSchema fs3 =new ResourceFieldSchema(c3);            

rs.setFields(new ResourceFieldSchema[]{fs1,fs2,fs3});            

return rs;

}

 

这一个简单的例子中就返回了直接使用模式的形式 

grunt> register  /home/pig/pig-0.11.0/udflib/myStorage.jar            

grunt> a = load 'student' using com.hcr.hadoop.pig.MyStorage(',','/');

grunt> describe a; a: {c1: int,c2: int,c3: double}

grunt> b = foreach  a generate c1,c2,c3;

grunt> describe b;

b: {c1: int,c2: int,c3: double}

 

摘录地址：http://blog.csdn.net/ruishenh/article/details/12192391

转载于:https://www.cnblogs.com/anny-1980/p/3673569.html