def test_nrt():
indexer = engine.Indexer(nrt=True)
indexer.add()
assert indexer.count() == 0 and not indexer.current
indexer.refresh()
assert indexer.count() == 1 and indexer.current
searcher = engine.IndexSearcher(indexer.directory)
assert searcher.count() == 0 and searcher.current
indexer.add()
indexer.commit()
assert indexer.count() == engine.IndexSearcher(
indexer.directory).count() == 2
def __init__(self, *directories, **kwargs):
if len(directories) > 1:
self.searcher = engine.MultiSearcher(directories, **kwargs)
else:
self.searcher = engine.IndexSearcher(*directories, **kwargs)
self.updated = time.time()
self.query_map = {}
def test_indexes(tempdir):
with pytest.raises(TypeError):
engine.IndexSearcher()
with pytest.raises(lucene.JavaError):
engine.Indexer(tempdir, 'r')
indexer = engine.Indexer()
indexer.set('name', engine.Field.String, stored=True)
indexer.set('text', engine.Field.Text)
with engine.Indexer(tempdir) as temp:
temp.add()
with pytest.raises(KeyError), engine.Indexer(tempdir) as temp:
temp.add()
temp.add(missing='')
for other in (temp, temp.directory, tempdir):
indexer += other
assert len(indexer) == 3
analyzer = engine.Analyzer.whitespace()
indexer.add(text=analyzer.tokens('?'), name=util.BytesRef('{}'))
indexer.commit()
assert indexer[next(indexer.docs('text', '?'))]['name'] == '{}'
indexer.delete('text', '?')
indexer.commit(merge=True)
assert not indexer.hasDeletions()
indexer.commit(merge=1)
assert len(list(indexer.readers)) == 1
reader = engine.indexers.IndexReader(indexer.indexReader)
del reader.indexReader
with pytest.raises(AttributeError):
reader.maxDoc
del indexer.indexSearcher
with pytest.raises(AttributeError):
indexer.search
indexer = engine.Indexer(tempdir)
indexer.add()
indexer.commit()
files = set(os.listdir(tempdir))
path = os.path.join(tempdir, 'temp')
with indexer.snapshot() as commit:
indexer.commit(merge=1)
assert indexer.indexCommit.generation > commit.generation
engine.indexers.copy(commit, path)
assert set(os.listdir(path)) == set(commit.fileNames) < files < set(
os.listdir(tempdir))
filepath = os.path.join(path, commit.segmentsFileName)
os.remove(filepath)
open(filepath, 'w').close()
with pytest.raises(OSError):
engine.indexers.copy(commit, path)
with pytest.raises(lucene.JavaError):
indexer.check(tempdir)
del indexer
assert engine.Indexer(tempdir)
assert not os.path.exists(os.path.join(tempdir, commit.segmentsFileName))
assert engine.IndexWriter.check(tempdir).clean
assert not engine.IndexWriter.check(tempdir, fix=True).numBadSegments
def __init__(self, *directories, **kwargs):
self.urls = collections.deque(kwargs.pop('urls', ()))
if self.urls:
engine.IndexWriter(*directories).close()
self.searcher = engine.MultiSearcher(
directories, **
kwargs) if len(directories) > 1 else engine.IndexSearcher(
*directories, **kwargs)
self.updated = time.time()
self.query_map = {}
def test_grouping(tempdir, indexer, zipcodes):
field = indexer.fields['location'] = engine.NestedField(
'state.county.city', docValuesType='sorted')
for doc in zipcodes:
if doc['state'] in ('CA', 'AK', 'WY', 'PR'):
lat, lng = ('{0:08.3f}'.format(doc.pop(l))
for l in ['latitude', 'longitude'])
location = '.'.join(doc[name]
for name in ['state', 'county', 'city'])
indexer.add(doc, latitude=lat, longitude=lng, location=location)
indexer.commit()
states = list(indexer.terms('state'))
assert states[0] == 'AK' and states[-1] == 'WY'
counties = [
term.split('.')[-1] for term in indexer.terms('state.county', 'CA')
]
hits = indexer.search(field.prefix('CA'))
assert sorted({hit['county'] for hit in hits}) == counties
assert counties[0] == 'Alameda' and counties[-1] == 'Yuba'
cities = [
term.split('.')[-1]
for term in indexer.terms('state.county.city', 'CA.Los Angeles')
]
hits = indexer.search(field.prefix('CA.Los Angeles'))
assert sorted({hit['city'] for hit in hits}) == cities
assert cities[0] == 'Acton' and cities[-1] == 'Woodland Hills'
(hit, ) = indexer.search('zipcode:90210')
assert hit['state'] == 'CA' and hit['county'] == 'Los Angeles' and hit[
'city'] == 'Beverly Hills' and hit['longitude'] == '-118.406'
query = Q.prefix('zipcode', '90')
((field, facets), ) = indexer.facets(query, 'state.county').items()
assert field == 'state.county'
la, orange = sorted(filter(facets.get, facets))
assert la == 'CA.Los Angeles' and facets[la] > 100
assert orange == 'CA.Orange' and facets[orange] > 10
queries = {
term: Q.term(field, term)
for term in indexer.terms(field, 'CA.')
}
((field, facets), ) = indexer.facets(query, **{field: queries}).items()
assert all(value.startswith('CA.')
for value in facets) and set(facets) == set(queries)
assert facets['CA.Los Angeles'] == 264
groups = indexer.groupby(field, Q.term('state', 'CA'), count=1)
assert len(groups) == 1 < groups.count
(hits, ) = groups
assert hits.value == 'CA.Los Angeles' and len(
hits) == 1 and hits.count > 100
grouping = engine.documents.GroupingSearch(field,
sort=search.Sort(
indexer.sortfield(field)),
cache=False,
allGroups=True)
assert all(
grouping.search(indexer.indexSearcher, Q.alldocs()).facets.values())
assert len(grouping) == len(list(grouping)) > 100
assert set(grouping) > set(facets)
hits = indexer.search(query, timeout=-1)
assert not hits and not hits.count and math.isnan(hits.maxscore)
hits = indexer.search(query, timeout=10)
assert len(hits) == hits.count == indexer.count(
query) and hits.maxscore == 1.0
directory = store.RAMDirectory()
query = Q.term('state', 'CA')
size = indexer.copy(directory, query)
searcher = engine.IndexSearcher(directory)
assert len(searcher) == size and list(searcher.terms('state')) == ['CA']
path = os.path.join(tempdir, 'temp')
size = indexer.copy(path, exclude=query, merge=1)
assert len(searcher) + size == len(indexer)
searcher = engine.IndexSearcher(path)
assert len(searcher.segments) == 1 and 'CA' not in searcher.terms('state')
directory.close()