Package com.twitter.elephantbird.mapreduce.input.LuceneIndexInputFormat

Examples of com.twitter.elephantbird.mapreduce.input.LuceneIndexInputFormat.LuceneIndexInputSplit

120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
    assertTrue(split.getIndexDirs().get(0).toString().endsWith("sample_indexes/index-2"));
  }

  @Test
  public void testLuceneIndexInputSplit() throws Exception {
    LuceneIndexInputSplit orig = new LuceneIndexInputSplit(
        Lists.newArrayList(new Path("/index/test"),
                           new Path("/index/test2"),
                           new Path("/index/test3")), 500L);

    ByteArrayOutputStream bytesOut = new ByteArrayOutputStream();
    DataOutputStream dataOut = new DataOutputStream(bytesOut);

    orig.write(dataOut);

    LuceneIndexInputSplit deSerialized = new LuceneIndexInputSplit();
    deSerialized.readFields(new DataInputStream((new ByteArrayInputStream(bytesOut.toByteArray()))));

    assertEquals(orig.getIndexDirs(), deSerialized.getIndexDirs());
    assertEquals(orig.getLength(), deSerialized.getLength());

    assertEquals(0, orig.compareTo(deSerialized));

    LuceneIndexInputSplit smaller = new LuceneIndexInputSplit(
        Lists.newArrayList(new Path("/index/small")), 100L);

    assertTrue(orig.compareTo(smaller) > 0);
    assertTrue(smaller.compareTo(orig) < 0);
  }
View Full Code Here
155
156
157
158
159
160
161
162
163
164
165
    String[] paths = new String[] {
        "/index/1", "/index/2", "/index/3", "/index/4", "/index/5", "/index/6"};
    Long[] sizes = new Long[]{500L, 300L, 100L, 150L, 1200L, 500L};

    for (int i = 0; i < paths.length; i++) {
      splits.add(new LuceneIndexInputSplit(Lists.newArrayList(new Path(paths[i])), sizes[i]));
    }

    List<InputSplit> combined = lif.combineSplits(splits, 1000L, 10000L);
    assertEquals(3, combined.size());
View Full Code Here
186
187
188
189
190
191
192
193
194
195
196
  @Test
  public void testCombineSplitsOneSplit() throws Exception {
    DummyLuceneInputFormat lif = new DummyLuceneInputFormat();

    PriorityQueue<LuceneIndexInputSplit> splits = new PriorityQueue<LuceneIndexInputSplit>();
    splits.add(new LuceneIndexInputSplit(Lists.newArrayList(new Path("/index/1")), 1500L));

    List<InputSplit> combined = lif.combineSplits(splits, 1000L, 10000L);
    assertEquals(1, combined.size());

    List<Path> dirs = ((LuceneIndexInputSplit) combined.get(0)).getIndexDirs();
View Full Code Here
206
207
208
209
210
211
212
213
214
215
216
    PriorityQueue<LuceneIndexInputSplit> splits = new PriorityQueue<LuceneIndexInputSplit>();
    String[] paths = new String[]{"/index/1", "/index/2", "/index/3"};
    Long[] sizes = new Long[]{1500L, 1501L, 1502L};
    for (int i = 0; i < paths.length; i++) {
      splits.add(new LuceneIndexInputSplit(Lists.newArrayList(new Path(paths[i])), sizes[i]));
    }

    List<InputSplit> combined = lif.combineSplits(splits, 1000L, 10000L);
    assertEquals(3, combined.size());
View Full Code Here
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
          break;
        default:
          sizes[i] = 1L;
          paths[i] = "/index/small-" + i;
      }
      splits.add(new LuceneIndexInputSplit(Lists.newArrayList(new Path(paths[i])), sizes[i]));
    }

    List<InputSplit> combined = lif.combineSplits(splits, 150L, 10L);
    assertEquals(12, combined.size());

    for (int i = 0; i < 9; i++) {
      LuceneIndexInputSplit split = (LuceneIndexInputSplit) combined.get(i);
      assertEquals(10L, split.getIndexDirs().size());
      assertEquals(10L, split.getLength());
      for (Path p : split.getIndexDirs()) {
        assertTrue(p.toString().startsWith("/index/small-"));
      }
    }

    LuceneIndexInputSplit split = (LuceneIndexInputSplit) combined.get(9);
    assertEquals(8, split.getIndexDirs().size());
    assertEquals(107, split.getLength());
    for (int i = 0; i < 7; i++) {
      assertTrue(split.getIndexDirs().get(i).toString().startsWith("/index/small-"));
    }
    assertEquals("/index/100", split.getIndexDirs().get(7).toString());

    split = (LuceneIndexInputSplit) combined.get(10);
    assertEquals(1, split.getIndexDirs().size());
    assertEquals(300, split.getLength());
    assertEquals("/index/300", split.getIndexDirs().get(0).toString());

    split = (LuceneIndexInputSplit) combined.get(11);
    assertEquals(1, split.getIndexDirs().size());
    assertEquals(500, split.getLength());
    assertEquals("/index/500", split.getIndexDirs().get(0).toString());
  }
View Full Code Here
142
143
144
145
146
147
148
149
150
151
152
153
  private void testLuceneIndexRecordReader(ArrayList<String> queryStrings,
      ArrayList<Path> indexPaths,
      ArrayList<ArrayList<ArrayList<Integer>>> indexesQueriesDocIds)
      throws Exception {

    LuceneIndexInputSplit split = createStrictMock(LuceneIndexInputSplit.class);
    expect(split.getIndexDirs()).andReturn(indexPaths);
    replay(split);

    Configuration conf = new Configuration();
    TaskAttemptContext context = createStrictMock(TaskAttemptContext.class);
    expect(HadoopCompat.getConfiguration(context)).andStubReturn(conf);
View Full Code Here
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
    LuceneIndexInputFormat.setInputPaths(inputPaths, conf);

    lif.loadConfig(conf);
    PriorityQueue<LuceneIndexInputSplit> splits = lif.findSplits(conf);
    LuceneIndexInputSplit split;
    split = splits.poll();
    assertEquals(4, split.getLength());
    assertTrue(split.getIndexDirs().get(0).toString().endsWith("sample_indexes/index-1"));

    split = splits.poll();
    assertEquals(6, split.getLength());
    assertTrue(split.getIndexDirs().get(0).toString().endsWith("sample_indexes/more-indexes/index-3"));

    split = splits.poll();
    assertEquals(20, split.getLength());
    assertTrue(split.getIndexDirs().get(0).toString().endsWith("sample_indexes/index-2"));

    assertTrue(splits.isEmpty());
  }
View Full Code Here
108
109
110
111
112
113
114
115
116
117
118
119
120
121
    LuceneIndexInputFormat.setMaxCombinedIndexSizePerSplitBytes(15L, conf);
    JobContext jobContext = createStrictMock(JobContext.class);
    expect(HadoopCompat.getConfiguration(jobContext)).andStubReturn(conf);
    replay(jobContext);
    List<InputSplit> splits = lif.getSplits(jobContext);
    LuceneIndexInputSplit split = (LuceneIndexInputSplit) splits.get(0);
    assertEquals(2, split.getIndexDirs().size());
    assertTrue(split.getIndexDirs().get(0).toString().endsWith("sample_indexes/index-1"));
    assertTrue(split.getIndexDirs().get(1).toString()
        .endsWith("sample_indexes/more-indexes/index-3"));
    split = (LuceneIndexInputSplit) splits.get(1);
    assertEquals(1, split.getIndexDirs().size());
    assertTrue(split.getIndexDirs().get(0).toString().endsWith("sample_indexes/index-2"));
  }
View Full Code Here
TOP

Related Classes of com.twitter.elephantbird.mapreduce.input.LuceneIndexInputFormat.LuceneIndexInputSplit

  • com.twitter.elephantbird.mapreduce.input.TestLuceneIndexInputFormat
  • com.twitter.elephantbird.mapreduce.input.TestLuceneIndexRecordReader
    • Copyright © 2018 www.massapicom. All rights reserved.
    All source code are property of their respective owners. Java is a trademark of Sun Microsystems, Inc and owned by ORACLE Inc. Contact coftware#gmail.com.