Package org.apache.lucene.util

Examples of org.apache.lucene.util.OpenBitSet.cardinality()

67
68
69
70
71
72
73
74
75
76
77
    IndexReader reader = IndexReader.open(rd);
   
    TermsFilter tf=new TermsFilter();
    tf.addTerm(new Term(fieldName,"19"));
    OpenBitSet bits = (OpenBitSet)tf.getDocIdSet(reader);
    assertEquals("Must match nothing", 0, bits.cardinality());

    tf.addTerm(new Term(fieldName,"20"));
    bits = (OpenBitSet)tf.getDocIdSet(reader);
    assertEquals("Must match 1", 1, bits.cardinality());
   
View Full Code Here
71
72
73
74
75
76
77
78
79
80
81
    OpenBitSet bits = (OpenBitSet)tf.getDocIdSet(reader);
    assertEquals("Must match nothing", 0, bits.cardinality());

    tf.addTerm(new Term(fieldName,"20"));
    bits = (OpenBitSet)tf.getDocIdSet(reader);
    assertEquals("Must match 1", 1, bits.cardinality());
   
    tf.addTerm(new Term(fieldName,"10"));
    bits = (OpenBitSet)tf.getDocIdSet(reader);
    assertEquals("Must match 2", 2, bits.cardinality());
   
View Full Code Here
75
76
77
78
79
80
81
82
83
84
85
    bits = (OpenBitSet)tf.getDocIdSet(reader);
    assertEquals("Must match 1", 1, bits.cardinality());
   
    tf.addTerm(new Term(fieldName,"10"));
    bits = (OpenBitSet)tf.getDocIdSet(reader);
    assertEquals("Must match 2", 2, bits.cardinality());
   
    tf.addTerm(new Term(fieldName,"00"));
    bits = (OpenBitSet)tf.getDocIdSet(reader);
    assertEquals("Must match 2", 2, bits.cardinality());
       
View Full Code Here
79
80
81
82
83
84
85
86
87
    bits = (OpenBitSet)tf.getDocIdSet(reader);
    assertEquals("Must match 2", 2, bits.cardinality());
   
    tf.addTerm(new Term(fieldName,"00"));
    bits = (OpenBitSet)tf.getDocIdSet(reader);
    assertEquals("Must match 2", 2, bits.cardinality());
       
  }
}
View Full Code Here
205
206
207
208
209
210
211
212
213
214
215
        }
      }
      // AND the term's bitset with cluster doc bitset to get the term's in-cluster frequency.
      // This modifies the termBitset, but that's fine as we are not using it anywhere else.
      termBitset.and(clusterDocBitset);
      int inclusterDF = (int) termBitset.cardinality();

      TermEntry entry = new TermEntry(term.utf8ToString(), count++, inclusterDF);
      termEntryMap.put(entry.getTerm(), entry);

    }
View Full Code Here
261
262
263
264
265
266
267
268
269
270
271
      }
      if (idSet.contains(id)) {
        bitset.set(i);
      }
    }
    log.info("Created bitset for in-cluster documents : {}", bitset.cardinality());
    return bitset;
  }

  private static double scoreDocumentFrequencies(long inDF, long outDF, long clusterSize, long corpusSize) {
    long k12 = clusterSize - inDF;
View Full Code Here
105
106
107
108
109
110
111
112
113
114
115
        return result;
    }

    public long cardinality(IndexReader reader, Term surfaceFormTerm) throws IOException {
        OpenBitSet surfaceFormDocIdSet = (OpenBitSet) getDocIdSet(reader, surfaceFormTerm);
        return surfaceFormDocIdSet.cardinality();
    }

    /**
     * Returns the number of documents containing the surface form contextTerm AND the contextTerm provided.
     * @param reader
View Full Code Here
202
203
204
205
206
207
208
209
210
211
212
        }

        // AND the term's bitset with cluster doc bitset to get the term's in-cluster frequency.
        // This modifies the termBitset, but that's fine as we are not using it anywhere else.
        termBitset.and(clusterDocBitset);
        int inclusterDF = (int) termBitset.cardinality();

        TermEntry entry = new TermEntry(term.text(), count++, inclusterDF);
        termEntryMap.put(entry.getTerm(), entry);
      } while (te.next());
    } finally {
View Full Code Here
255
256
257
258
259
260
261
262
263
264
265
      }
      if (idSet.contains(id)) {
        bitset.set(i);
      }
    }
    log.info("Created bitset for in-cluster documents : {}", bitset.cardinality());
    return bitset;
  }

  private static double scoreDocumentFrequencies(long inDF, long outDF, long clusterSize, long corpusSize) {
    long k12 = clusterSize - inDF;
View Full Code Here
192
193
194
195
196
197
198
199
200
201
202
          break;
        }
      } while (termEnum.next());
     
      // sort
      this.numberOfDocsWithField = docsWithFieldBitSet.cardinality();
      if (Val.chkStr(request.getSortBy()).equalsIgnoreCase("name")) {
        this.termAccumulator.sortByName();
      } else {
        this.termAccumulator.sortByFrequency();
      }
View Full Code Here
TOP
Copyright © 2018 www.massapi.com. 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.