Package org.apache.nutch.searcher.Query

Examples of org.apache.nutch.searcher.Query.Clause

58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
    // examine each clause in the Nutch query
    Clause[] clauses = input.getClauses();
   
    for (int i = 0; i <clauses.length; i++) {
      Clause c = clauses[i];
     
      //skip if not date clauses
      if (!c.getField().equals(FIELD_NAME))
        continue;
           
      String x = c.getTerm().toString();
      
      Matcher matcher = pattern.matcher(x);
      if (!matcher.matches()) {
        throw new QueryException("Wrong query syntax "+FIELD_NAME+":"+x);
      }

      // do it as lucene RangeQuery
      Term xLower = new Term(FIELD_NAME, matcher.group(1));
      Term xUpper = new Term(FIELD_NAME, matcher.group(2));

      // inclusive
      RangeQuery rangeQuery = new RangeQuery(xLower, xUpper, true);

      rangeQuery.setBoost(0.0f);                  // trigger filterization
         
      output.add(rangeQuery,
          (c.isProhibited()
              ? BooleanClause.Occur.MUST_NOT
              : (c.isRequired()
                  ? BooleanClause.Occur.MUST
                  : BooleanClause.Occur.SHOULD
                 )
           ));
            
View Full Code Here
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
    throws QueryException {
   
    // examine each clause in the Nutch query
    Clause[] clauses = input.getClauses();
    for (int i = 0; i < clauses.length; i++) {
      Clause c = clauses[i];

      // skip non-matching clauses
      if (!c.getField().equals(field))
        continue;

      // get the field value from the clause
      // raw fields are guaranteed to be Terms, not Phrases
      String value = c.getTerm().toString();
      if (lowerCase)
        value = value.toLowerCase();

      // add a Lucene TermQuery for this clause
      TermQuery clause = new TermQuery(new Term(field, value));
      // set boost
      clause.setBoost(boost);
      // add it as specified in query
     
      output.add(clause,
          (c.isProhibited()
              ? BooleanClause.Occur.MUST_NOT
              : (c.isRequired()
                  ? BooleanClause.Occur.MUST
                  : BooleanClause.Occur.SHOULD
                 )
           ));
    }
View Full Code Here
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
    throws QueryException {
   
    // examine each clause in the Nutch query
    Clause[] clauses = input.getClauses();
    for (int i = 0; i < clauses.length; i++) {
      Clause c = clauses[i];

      // skip non-matching clauses
      if (!c.getField().equals(field))
        continue;

      // get the field value from the clause
      // raw fields are guaranteed to be Terms, not Phrases
      String value = c.getTerm().toString();
      if (lowerCase)
        value = value.toLowerCase();

      // add a Lucene TermQuery for this clause
      TermQuery clause = new TermQuery(new Term(field, value));
      // set boost
      clause.setBoost(boost);
      // add it as specified in query
      output.add(clause, c.isRequired(), c.isProhibited());
    }
   
    // return the modified Lucene query
    return output;
  }
View Full Code Here
77
78
79
80
81
82
83
84
85
86
87
88
89
  /** Run all defined filters. */
  public static BooleanQuery filter(Query input) throws QueryException {
    // first check that all field names are claimed by some plugin
    Clause[] clauses = input.getClauses();
    for (int i = 0; i < clauses.length; i++) {
      Clause c = clauses[i];
      if (!isField(c.getField()))
        throw new QueryException("Not a known field name:"+c.getField());
    }

    // then run each plugin
    BooleanQuery output = new BooleanQuery();
    for (int i = 0 ; i < CACHE.length; i++) {
View Full Code Here
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
    // examine each clause in the Nutch query
    Clause[] clauses = input.getClauses();
   
    for (int i = 0; i <clauses.length; i++) {
      Clause c = clauses[i];
     
      //skip if not date clauses
      if (!c.getField().equals(FIELD_NAME))
        continue;
           
      String x = c.getTerm().toString();
      
      Matcher matcher = pattern.matcher(x);
      if (!matcher.matches()) {
        throw new QueryException("Wrong query syntax "+FIELD_NAME+":"+x);
      }

      // do it as lucene RangeQuery
      Term xLower = new Term(FIELD_NAME, matcher.group(1));
      Term xUpper = new Term(FIELD_NAME, matcher.group(2));

      // inclusive
      RangeQuery rangeQuery = new RangeQuery(xLower, xUpper, true);

      rangeQuery.setBoost(0.0f);                  // trigger filterization
         
      output.add(rangeQuery, c.isRequired(), c.isProhibited());
            
    }

    return output;
  }
View Full Code Here
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
    throws QueryException {
   
    // examine each clause in the Nutch query
    Clause[] clauses = input.getClauses();
    for (int i = 0; i < clauses.length; i++) {
      Clause c = clauses[i];

      // skip non-matching clauses
      if (!c.getField().equals(field))
        continue;

      // optimize phrase clause
      if (c.isPhrase()) {
        String[] opt = CommonGrams.optimizePhrase(c.getPhrase(), field);
        if (opt.length==1) {
          c = new Clause(new Query.Term(opt[0]),
                         c.isRequired(), c.isProhibited());
        } else {
          c = new Clause(new Phrase(opt), c.isRequired(), c.isProhibited());
        }
      }

      // construct appropriate Lucene clause
      org.apache.lucene.search.Query luceneClause;
      if (c.isPhrase()) {
        Phrase nutchPhrase = c.getPhrase();
        Query.Term[] terms = nutchPhrase.getTerms();
        PhraseQuery lucenePhrase = new PhraseQuery();
        for (int j = 0; j < terms.length; j++) {
          lucenePhrase.add(new Term(field, terms[j].toString()));
        }
        luceneClause = lucenePhrase;
      } else {
        luceneClause = new TermQuery(new Term(field, c.getTerm().toString()));
      }

      // set boost
      luceneClause.setBoost(boost);
      // add it as specified in query
      output.add(luceneClause, c.isRequired(), c.isProhibited());
    }
   
    // return the modified Lucene query
    return output;
  }
View Full Code Here
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
    // examine each clause in the Nutch query
    Clause[] clauses = input.getClauses();
   
    for (int i = 0; i <clauses.length; i++) {
      Clause c = clauses[i];
     
      //skip if not date clauses
      if (!c.getField().equals(FIELD_NAME))
        continue;
           
      String x = c.getTerm().toString();
      
      Matcher matcher = pattern.matcher(x);
      if (!matcher.matches()) {
        throw new QueryException("Wrong query syntax "+FIELD_NAME+":"+x);
      }

      // do it as lucene RangeQuery
      String xLower = matcher.group(1);
      String xUpper = matcher.group(2);

      // inclusive
      TermRangeQuery rangeQuery = new TermRangeQuery(
          c.getField(), xLower, xUpper, true, true);

      rangeQuery.setBoost(0.0f);                  // trigger filterization
         
      output.add(rangeQuery,
          (c.isProhibited()
              ? BooleanClause.Occur.MUST_NOT
              : (c.isRequired()
                  ? BooleanClause.Occur.MUST
                  : BooleanClause.Occur.SHOULD
                 )
           ));
            
View Full Code Here
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
    // examine each clause in the Nutch query
    Clause[] clauses = input.getClauses();
   
    for (int i = 0; i <clauses.length; i++) {
      Clause c = clauses[i];
     
      //skip if not date clauses
      if (!c.getField().equals(FIELD_NAME))
        continue;
           
      String x = c.getTerm().toString();
      
      Matcher matcher = pattern.matcher(x);
      if (!matcher.matches()) {
        throw new QueryException("Wrong query syntax "+FIELD_NAME+":"+x);
      }

      // do it as lucene RangeQuery
      Term xLower = new Term(FIELD_NAME, matcher.group(1));
      Term xUpper = new Term(FIELD_NAME, matcher.group(2));

      // inclusive
      RangeQuery rangeQuery = new RangeQuery(xLower, xUpper, true);

      rangeQuery.setBoost(0.0f);                  // trigger filterization
         
      output.add(rangeQuery,
          (c.isProhibited()
              ? BooleanClause.Occur.MUST_NOT
              : (c.isRequired()
                  ? BooleanClause.Occur.MUST
                  : BooleanClause.Occur.SHOULD
                 )
           ));
            
View Full Code Here
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
    throws QueryException {
   
    // examine each clause in the Nutch query
    Clause[] clauses = input.getClauses();
    for (int i = 0; i < clauses.length; i++) {
      Clause c = clauses[i];

      // skip non-matching clauses
      if (!c.getField().equals(field))
        continue;

      // optimize phrase clause
      if (c.isPhrase()) {
        String[] opt = this.commonGrams.optimizePhrase(c.getPhrase(), field);
        if (opt.length==1) {
          c = new Clause(new Query.Term(opt[0]),
                         c.isRequired(), c.isProhibited(), getConf());
        } else {
          c = new Clause(new Phrase(opt), c.isRequired(), c.isProhibited(), getConf());
        }
      }

      // construct appropriate Lucene clause
      org.apache.lucene.search.Query luceneClause;
      if (c.isPhrase()) {
        Phrase nutchPhrase = c.getPhrase();
        Query.Term[] terms = nutchPhrase.getTerms();
        PhraseQuery lucenePhrase = new PhraseQuery();
        for (int j = 0; j < terms.length; j++) {
          lucenePhrase.add(new Term(field, terms[j].toString()));
        }
        luceneClause = lucenePhrase;
      } else {
        luceneClause = new TermQuery(new Term(field, c.getTerm().toString()));
      }

      // set boost
      luceneClause.setBoost(boost);
      // add it as specified in query
     
      output.add(luceneClause,
          (c.isProhibited()
              ? BooleanClause.Occur.MUST_NOT
              : (c.isRequired()
                  ? BooleanClause.Occur.MUST
                  : BooleanClause.Occur.SHOULD
                 )
           ));
    }
View Full Code Here
95
96
97
98
99
100
101
102
103
104
105
106
107
  /** Run all defined filters. */
  public BooleanQuery filter(Query input) throws QueryException {
    // first check that all field names are claimed by some plugin
    Clause[] clauses = input.getClauses();
    for (int i = 0; i < clauses.length; i++) {
      Clause c = clauses[i];
      if (!isField(c.getField()))
        throw new QueryException("Not a known field name:"+c.getField());
    }

    // then run each plugin
    BooleanQuery output = new BooleanQuery();
    for (int i = 0; i < this.queryFilters.length; i++) {
View Full Code Here
TOP

Related Classes of org.apache.nutch.searcher.Query.Clause

  • org.apache.nutch.searcher.custom.CustomFieldQueryFilter
  • org.apache.nutch.searcher.FieldQueryFilter
  • org.apache.nutch.searcher.more.DateQueryFilter
  • org.apache.nutch.searcher.QueryFilters
  • org.apache.nutch.searcher.RawFieldQueryFilter
  • org.archive.access.nutch.searcher.WaxDateClosestQueryFilter
  • org.archive.access.nutch.searcher.WaxDateQueryFilter
  • org.archive.access.nutch.searcher.WaxExacturlExpandQueryFilter
  • org.archive.nutch.trec.TRECFeaturesExtractor
    • 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.