def _getFullStrings(self):
mergePolicy = LogDocMergePolicy()
mergePolicy.setMergeFactor(97)
directory = RAMDirectory()
self.dirs.append(directory)
writer = self.getWriter(directory=directory,
analyzer=SimpleAnalyzer(Version.LUCENE_CURRENT),
maxBufferedDocs=4, mergePolicy=mergePolicy)
onlyStored = FieldType()
onlyStored.setStored(True)
fixedLen = self.getRandomNumber(2, 8)
fixedLen2 = self.getRandomNumber(1, 4)
for i in xrange(NUM_STRINGS):
doc = Document()
num = self.getRandomCharString(self.getRandomNumber(2, 8), 48, 52)
doc.add(Field("tracer", num, onlyStored))
doc.add(StringField("string", num, Field.Store.NO))
if self.supportsDocValues:
if self.dvStringSorted:
doc.add(SortedDocValuesField("string_dv", BytesRef(num)))
else:
doc.add(BinaryDocValuesField("string_dv", BytesRef(num)))
num2 = self.getRandomCharString(self.getRandomNumber(1, 4), 48, 50)
doc.add(StringField("string2", num2, Field.Store.NO))
if self.supportsDocValues:
if self.dvStringSorted:
doc.add(SortedDocValuesField("string2_dv", BytesRef(num2)))
else:
doc.add(BinaryDocValuesField("string2_dv", BytesRef(num2)))
doc.add(Field("tracer2", num2, onlyStored))
for f2 in doc.getFields():
if f2.fieldType().indexed() and not f2.fieldType().omitNorms():
Field.cast_(f2).setBoost(2.0)
numFixed = self.getRandomCharString(fixedLen, 48, 52)
doc.add(Field("fixed_tracer", numFixed, onlyStored))
doc.add(StringField("string_fixed", numFixed, Field.Store.NO))
if self.supportsDocValues:
if self.dvStringSorted:
doc.add(SortedDocValuesField("string_fixed_dv", BytesRef(numFixed)))
else:
doc.add(BinaryDocValuesField("string_fixed_dv", BytesRef(numFixed)))
num2Fixed = self.getRandomCharString(fixedLen2, 48, 52)
doc.add(StringField("string2_fixed", num2Fixed, Field.Store.NO))
if self.supportsDocValues:
if self.dvStringSorted:
doc.add(SortedDocValuesField("string2_fixed_dv", BytesRef(num2Fixed)))
else:
doc.add(BinaryDocValuesField("string2_fixed_dv", BytesRef(num2Fixed)))
doc.add(Field("tracer2_fixed", num2Fixed, onlyStored))
for f2 in doc.getFields():
if f2.fieldType().indexed() and not f2.fieldType().omitNorms():
Field.cast_(f2).setBoost(2.0)
writer.addDocument(doc)
writer.close()
return self.getSearcher(directory=directory)
def _getIndex(self, even, odd):
mergePolicy = LogDocMergePolicy()
mergePolicy.setMergeFactor(1000)
directory = RAMDirectory()
self.dirs.append(directory)
writer = self.getWriter(directory=directory,
analyzer=SimpleAnalyzer(Version.LUCENE_CURRENT),
maxBufferedDocs=2, mergePolicy=mergePolicy)
if self.dvStringSorted:
# Index sorted
stringDVType = FieldInfo.DocValuesType.SORTED
elif self.notSorted:
# Index non-sorted
stringDVType = FieldInfo.DocValuesType.BINARY
else:
# sorted anyway
stringDVType = FieldInfo.DocValuesType.SORTED
ft1 = FieldType()
ft1.setStored(True)
ft2 = FieldType()
ft2.setIndexed(True)
for i in xrange(len(self.data)):
if (i % 2 == 0 and even) or (i % 2 == 1 and odd):
doc = Document()
doc.add(Field("tracer", self.data[i][0], ft1))
doc.add(TextField("contents", self.data[i][1], Field.Store.NO))
if self.data[i][2] is not None:
doc.add(StringField("int", self.data[i][2], Field.Store.NO))
if self.supportsDocValues:
doc.add(NumericDocValuesField("int_dv", Long.parseLong(self.data[i][2])))
if self.data[i][3] is not None:
doc.add(StringField("float", self.data[i][3], Field.Store.NO))
if self.supportsDocValues:
doc.add(FloatDocValuesField("float_dv", Float.parseFloat(self.data[i][3])))
if self.data[i][4] is not None:
doc.add(StringField("string", self.data[i][4], Field.Store.NO))
if self.supportsDocValues:
if stringDVType == FieldInfo.DocValuesType.SORTED:
doc.add(SortedDocValuesField("string_dv", BytesRef(self.data[i][4])))
elif stringDVType == FieldInfo.DocValuesType.BINARY:
doc.add(BinaryDocValuesField("string_dv", BytesRef(self.data[i][4])))
else:
raise ValueError("unknown type " + stringDVType)
if self.data[i][5] is not None:
doc.add(StringField("custom", self.data[i][5], Field.Store.NO))
if self.data[i][6] is not None:
doc.add(StringField("i18n", self.data[i][6], Field.Store.NO))
if self.data[i][7] is not None:
doc.add(StringField("long", self.data[i][7], Field.Store.NO))
if self.data[i][8] is not None:
doc.add(StringField("double", self.data[i][8], Field.Store.NO))
if self.supportsDocValues:
doc.add(NumericDocValuesField("double_dv", Double.doubleToRawLongBits(Double.parseDouble(self.data[i][8]))))
if self.data[i][9] is not None:
doc.add(StringField("short", self.data[i][9], Field.Store.NO))
if self.data[i][10] is not None:
doc.add(StringField("byte", self.data[i][10], Field.Store.NO))
if self.data[i][11] is not None:
doc.add(StringField("parser", self.data[i][11], Field.Store.NO))
for f in doc.getFields():
if f.fieldType().indexed() and not f.fieldType().omitNorms():
Field.cast_(f).setBoost(2.0)
writer.addDocument(doc)
reader = writer.getReader()
writer.close()
return self.getSearcher(reader=reader)
