test/test_Similarity.py

   1 # ====================================================================
   2 #   Licensed under the Apache License, Version 2.0 (the "License");
   3 #   you may not use this file except in compliance with the License.
   4 #   You may obtain a copy of the License at
   5 #
   6 #       http://www.apache.org/licenses/LICENSE-2.0
   7 #
   8 #   Unless required by applicable law or agreed to in writing, software
   9 #   distributed under the License is distributed on an "AS IS" BASIS,
  10 #   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
  11 #   See the License for the specific language governing permissions and
  12 #   limitations under the License.
  13 # ====================================================================
  14
  15 from unittest import TestCase, main
  16 from lucene import *
  17
  18
  19 class SimpleSimilarity(PythonSimilarity):
  20
  21     def computeNorm(self, field, state):
  22         return 1.0
  23
  24     def queryNorm(self, sumOfSquaredWeights):
  25         return 1.0
  26
  27     def tf(self, freq):
  28         return freq
  29
  30     def sloppyFreq(self, distance):
  31         return 2.0
  32
  33     def idfTerms(self, terms, searcher):
  34         return 1.0
  35
  36     def idf(self, docFreq, numDocs):
  37         return 1.0
  38
  39     def coord(self, overlap, maxOverlap):
  40         return 1.0
  41
  42     def scorePayload(self, docId, fieldName, start, end, payload,
  43                      offset, length):
  44         return 1.0
  45
  46
  47 class SimilarityTestCase(TestCase):
  48     """
  49     Unit tests ported from Java Lucene
  50     """
  51
  52     def testSimilarity(self):
  53
  54         store = RAMDirectory()
  55         writer = IndexWriter(store, SimpleAnalyzer(), True,
  56                              IndexWriter.MaxFieldLength.LIMITED)
  57         writer.setSimilarity(SimpleSimilarity())
  58
  59         d1 = Document()
  60         d1.add(Field("field", "a c", Field.Store.YES, Field.Index.ANALYZED))
  61
  62         d2 = Document()
  63         d2.add(Field("field", "a b c", Field.Store.YES, Field.Index.ANALYZED))
  64
  65         writer.addDocument(d1)
  66         writer.addDocument(d2)
  67         writer.optimize()
  68         writer.close()
  69
  70         searcher = IndexSearcher(store, True)
  71         searcher.setSimilarity(SimpleSimilarity())
  72
  73         a = Term("field", "a")
  74         b = Term("field", "b")
  75         c = Term("field", "c")
  76
  77         class collector1(PythonCollector):
  78             def collect(_self, doc, score):
  79                 self.assertEqual(1.0, score)
  80             def setNextReader(_self, reader, docBase):
  81                 pass
  82             def acceptsDocsOutOfOrder(_self):
  83                 return True
  84
  85         searcher.search(TermQuery(b), collector1())
  86
  87
  88         bq = BooleanQuery()
  89         bq.add(TermQuery(a), BooleanClause.Occur.SHOULD)
  90         bq.add(TermQuery(b), BooleanClause.Occur.SHOULD)
  91
  92         class collector2(PythonCollector):
  93             def collect(_self, doc, score):
  94                 self.assertEqual(doc + _self.base + 1, score)
  95             def setNextReader(_self, reader, docBase):
  96                 _self.base = docBase
  97             def acceptsDocsOutOfOrder(_self):
  98                 return True
  99
 100         searcher.search(bq, collector2())
 101
 102
 103         pq = PhraseQuery()
 104         pq.add(a)
 105         pq.add(c)
 106
 107         class collector3(PythonCollector):
 108             def collect(_self, doc, score):
 109                 self.assertEqual(1.0, score)
 110             def setNextReader(_self, reader, docBase):
 111                 pass
 112             def acceptsDocsOutOfOrder(_self):
 113                 return True
 114
 115         searcher.search(pq, collector3())
 116
 117
 118         pq.setSlop(2)
 119
 120         class collector4(PythonCollector):
 121             def collect(_self, doc, score):
 122                 self.assertEqual(2.0, score)
 123             def setNextReader(_self, reader, docBase):
 124                 pass
 125             def acceptsDocsOutOfOrder(_self):
 126                 return True
 127
 128         searcher.search(pq, collector4())
 129
 130
 131 if __name__ == "__main__":
 132     import sys, lucene
 133     lucene.initVM()
 134     if '-loop' in sys.argv:
 135         sys.argv.remove('-loop')
 136         while True:
 137             try:
 138                 main()
 139             except:
 140                 pass
 141     else:
 142          main()