def testIndexObject(self):
response = getData('add_response.txt')
# fake add response
output = fakehttp(self.mngr.getConnection(), response)
# indexing sends data
self.proc.index(Foo(id='500', name='python test doc'))
self.assertEqual(sortFields(str(output)), getData('add_request.txt'))
def testPartialIndexObject(self):
foo = Foo(id='500', name='foo', price=42.0)
# first index all attributes...
response = getData('add_response.txt')
output = fakehttp(self.mngr.getConnection(), response)
self.proc.index(foo)
self.assert_(str(output).find(
'<field name="price">42.0</field>') > 0, '"price" data not found')
# then only a subset...
response = getData('add_response.txt')
output = fakehttp(self.mngr.getConnection(), response)
self.proc.index(foo, attributes=['id', 'name'])
output = str(output)
self.assert_(
output.find('<field name="name">foo</field>') > 0,
'"name" data not found'
)
# at this point we'd normally check for a partial update:
# self.assertEqual(output.find('price'), -1, '"price" data found?')
# self.assertEqual(output.find('42'), -1, '"price" data found?')
# however, until SOLR-139 has been implemented (re)index operations
# always need to provide data for all attributes in the schema...
self.assert_(
output.find('<field name="price">42.0</field>') > 0,
'"price" data not found'
)
def test_commit_no_wait(self):
commit_request = getData('commit_request_no_wait.txt')
commit_response = getData('commit_response.txt')
c = SolrConnection(host='localhost:8983', persistent=True)
output = fakehttp(c, commit_response)
c.commit(waitFlush=False, waitSearcher=False)
self.failUnlessEqual(str(output), commit_request)
def testUnindexObject(self):
response = getData('delete_response.txt')
# fake response
output = fakehttp(self.mngr.getConnection(), response)
# unindexing sends data
self.proc.unindex(Foo(id='500', name='python test doc'))
self.assertEqual(str(output), getData('delete_request.txt'))
def test_optimize(self):
commit_request = getData("optimize_request.txt").rstrip(b"\n")
commit_response = getData("commit_response.txt")
c = SolrConnection(host="localhost:8983", persistent=True)
output = fakehttp(c, commit_response)
c.commit(optimize=True)
self.failUnlessEqual(str(output), commit_request.decode("utf-8"))
def test_commit_no_wait(self):
commit_request = getData('commit_request_no_wait.txt')
commit_response = getData('commit_response.txt')
c = SolrConnection(host='localhost:8983', persistent=True)
output = fakehttp(c, commit_response)
c.commit(waitFlush=False, waitSearcher=False)
self.failUnlessEqual(str(output), commit_request)
def testUnindexObject(self):
response = getData('delete_response.txt')
# fake response
output = fakehttp(self.mngr.getConnection(), response)
# unindexing sends data
self.proc.unindex(Foo(id='500', name='python test doc'))
self.assertEqual(str(output), getData('delete_request.txt'))
def testCommit(self):
response = getData('commit_response.txt')
# fake response
output = fakehttp(self.mngr.getConnection(), response)
# committing sends data
self.proc.commit()
self.assertEqual(str(output), getData('commit_request.txt'))
def testCommit(self):
response = getData('commit_response.txt')
# fake response
output = fakehttp(self.mngr.getConnection(), response)
# committing sends data
self.proc.commit()
self.assertEqual(str(output), getData('commit_request.txt'))
def test_add(self):
config = getConfig()
config.atomic_updates = True
add_request = getData('add_request.txt')
add_response = getData('add_response.txt')
c = SolrConnection(host='localhost:8983', persistent=True)
# fake schema response - caches the schema
fakehttp(c, getData('schema.xml'))
c.get_schema()
output = fakehttp(c, add_response)
c.add(id='500', name='python test doc')
res = c.flush()
self.assertEqual(len(res), 1) # one request was sent
res = res[0]
self.failUnlessEqual(str(output), add_request)
# Status
node = res.findall(".//int")[0]
self.failUnlessEqual(node.attrib['name'], 'status')
self.failUnlessEqual(node.text, '0')
# QTime
node = res.findall(".//int")[1]
self.failUnlessEqual(node.attrib['name'], 'QTime')
self.failUnlessEqual(node.text, '4')
res.find('QTime')
def test_add(self):
config = getConfig()
config.atomic_updates = True
add_request = getData("add_request.txt").rstrip(b"\n")
add_response = getData("add_response.txt")
c = SolrConnection(host="localhost:8983", persistent=True)
# fake schema response - caches the schema
fakehttp(c, getData("schema.xml"))
c.get_schema()
output = fakehttp(c, add_response)
c.add(id="500", name="python test doc")
res = c.flush()
self.assertEqual(len(res), 1) # one request was sent
res = res[0]
self.failUnlessEqual(str(output), add_request.decode("utf-8"))
# Status
node = res.findall(".//int")[0]
self.failUnlessEqual(node.attrib["name"], "status")
self.failUnlessEqual(node.text, "0")
# QTime
node = res.findall(".//int")[1]
self.failUnlessEqual(node.attrib["name"], "QTime")
self.failUnlessEqual(node.text, "4")
res.find("QTime")
def testIndexObject(self):
response = getData('add_response.txt')
output = fakehttp(self.mngr.getConnection(),
response) # fake add response
self.proc.index(Foo(id='500',
name='python test doc')) # indexing sends data
self.assertEqual(sortFields(str(output)), getData('add_request.txt'))
def test_search(self):
# XXX: Solr 7 has a new query param 'q.op' which can not be passed to
# the search method in Python.
# This is why we have commented out code here.
search_request = getData("search_request.txt").rstrip(b"\n")
search_request_py2 = getData("search_request_py2.txt").rstrip(b"\n")
search_response = getData("search_response.txt")
c = SolrConnection(host="localhost:8983", persistent=True)
output = fakehttp(c, search_response)
parameters = {
"q": "+id:[* TO *]",
"fl": "* score",
"wt": "xml",
"rows": "10",
"indent": "on",
"q.op": "AND",
"lowercaseOperators": "true",
"sow": "true",
}
res = c.search(**parameters)
res = fromstring(res.read())
normalize = lambda x: sorted(x.split(b"&")) # sort request params
self.assertIn(
normalize(output.get()),
[normalize(search_request),
normalize(search_request_py2)],
)
self.failUnless(res.find((".//doc")))
def testPartialIndexObject(self):
foo = Foo(id='500', name='foo', price=42.0)
# first index all attributes...
response = getData('add_response.txt')
output = fakehttp(self.mngr.getConnection(), response)
self.proc.index(foo)
self.assert_(
str(output).find('<field name="price">42.0</field>') > 0,
'"price" data not found')
# then only a subset...
response = getData('add_response.txt')
output = fakehttp(self.mngr.getConnection(), response)
self.proc.index(foo, attributes=['id', 'name'])
output = str(output)
self.assert_(
output.find('<field name="name">foo</field>') > 0,
'"name" data not found')
# at this point we'd normally check for a partial update:
# self.assertEqual(output.find('price'), -1, '"price" data found?')
# self.assertEqual(output.find('42'), -1, '"price" data found?')
# however, until SOLR-139 has been implemented (re)index operations
# always need to provide data for all attributes in the schema...
self.assert_(
output.find('<field name="price">42.0</field>') > 0,
'"price" data not found')
def test_optimize(self):
commit_request = getData('optimize_request.txt')
commit_response = getData('commit_response.txt')
c = SolrConnection(host='localhost:8983', persistent=True)
output = fakehttp(c, commit_response)
c.commit(optimize=True)
self.failUnlessEqual(str(output), commit_request)
def test_optimize(self):
commit_request = getData('optimize_request.txt')
commit_response = getData('commit_response.txt')
c = SolrConnection(host='localhost:8983', persistent=True)
output = fakehttp(c, commit_response)
c.commit(optimize=True)
self.failUnlessEqual(str(output), commit_request)
def test_commit_no_wait_flush(self):
commit_request = getData('commit_request.txt').rstrip('\n')
commit_response = getData('commit_response.txt')
c = SolrConnection(host='localhost:8983', persistent=True)
output = fakehttp(c, commit_response)
c.commit()
self.failUnlessEqual(str(output), commit_request)
def test_add(self):
config = getConfig()
config.atomic_updates = True
add_request = getData('add_request.txt').rstrip('\n')
add_response = getData('add_response.txt')
c = SolrConnection(host='localhost:8983', persistent=True)
# fake schema response - caches the schema
fakehttp(c, getData('schema.xml'))
c.get_schema()
output = fakehttp(c, add_response)
c.add(id='500', name='python test doc')
res = c.flush()
self.assertEqual(len(res), 1) # one request was sent
res = res[0]
self.failUnlessEqual(str(output), add_request)
# Status
node = res.findall(".//int")[0]
self.failUnlessEqual(node.attrib['name'], 'status')
self.failUnlessEqual(node.text, '0')
# QTime
node = res.findall(".//int")[1]
self.failUnlessEqual(node.attrib['name'], 'QTime')
self.failUnlessEqual(node.text, '4')
res.find('QTime')
def test_commit_no_wait_searcher(self):
commit_request = getData("commit_request_no_wait_searcher.txt").rstrip(
b"\n")
commit_response = getData("commit_response.txt")
c = SolrConnection(host="localhost:8983", persistent=True)
output = fakehttp(c, commit_response)
c.commit(waitSearcher=False)
self.failUnlessEqual(str(output), commit_request.decode("utf-8"))
def testIndexAccessorRaises(self):
response = getData('add_response.txt')
output = fakehttp(self.mngr.getConnection(), response) # fake add response
def brokenfunc():
raise ValueError
self.proc.index(Foo(id='500', name='python test doc',
text=brokenfunc)) # indexing sends data
self.assertEqual(sortFields(str(output)), getData('add_request.txt'))
def testCommit(self):
response = getData("commit_response.txt")
# fake response
output = fakehttp(self.mngr.getConnection(), response)
# committing sends data
self.proc.commit()
self.assertEqual(
str(output), getData("commit_request.txt").decode("utf-8").rstrip("\n")
)
def testUnindexObject(self):
response = getData("delete_response.txt")
# fake response
output = fakehttp(self.mngr.getConnection(), response)
# unindexing sends data
self.proc.unindex(Foo(id="500", name="python test doc"))
self.assertEqual(
str(output), getData("delete_request.txt").decode("utf-8").rstrip("\n")
)
def test_search(self):
search_request = getData('search_request.txt')
search_response = getData('search_response.txt')
c = SolrConnection(host='localhost:8983', persistent=True)
output = fakehttp(c, search_response)
res = c.search(q='+id:[* TO *]', wt='xml', rows='10', indent='on')
res = fromstring(res.read())
self.failUnlessEqual(str(output), search_request)
self.failUnless(res.find(('.//doc')))
def test_search(self):
search_request = getData('search_request.txt')
search_response = getData('search_response.txt')
c = SolrConnection(host='localhost:8983', persistent=True)
output = fakehttp(c, search_response)
res = c.search(q='+id:[* TO *]', wt='xml', rows='10', indent='on')
res = fromstring(res.read())
self.failUnlessEqual(str(output), search_request)
self.failUnless(res.find(('.//doc')))
def testIndexCatalogAwareObject(self):
"""Check that not only CMFCatalogAware objects are indexed but also
CatalogAware ones (i.e comments).
"""
response = getData('add_response.txt')
# fake add response
output = fakehttp(self.mngr.getConnection(), response)
# indexing sends data
self.proc.index(Comentish(id='500', name='python test doc'))
self.assertEqual(sortFields(str(output)), getData('add_request.txt'))
def testTwoRequests(self):
mngr = SolrConnectionManager(active=True)
proc = SolrIndexProcessor(mngr)
output = fakehttp(mngr.getConnection(), getData('schema.xml'),
getData('add_response.txt'))
proc.index(self.foo)
mngr.closeConnection()
self.assertEqual(len(output), 2)
self.failUnless(output.get().startswith(self.schema_request))
self.assertEqual(sortFields(output.get()), getData('add_request.txt'))
def testIndexObject(self):
response = getData("add_response.txt")
# fake add response
output = fakehttp(self.mngr.getConnection(), response)
# indexing sends data
self.proc.index(Foo(id="500", name="python test doc"))
self.assertEqual(
sortFields(str(output).encode("utf-8")),
getData("add_request.txt").rstrip(b"\n"),
)
def test_search(self):
search_request = getData('search_request.txt')
search_response = getData('search_response.txt')
c = SolrConnection(host='localhost:8983', persistent=True)
output = fakehttp(c, search_response)
res = c.search(q='+id:[* TO *]', fl='* score', wt='xml', rows='10', indent='on')
res = fromstring(res.read())
normalize = lambda x: sorted(x.split('&')) # sort request params
self.assertEqual(normalize(output.get()), normalize(search_request))
self.failUnless(res.find(('.//doc')))
def testTwoRequests(self):
mngr = SolrConnectionManager(active=True)
proc = SolrIndexProcessor(mngr)
output = fakehttp(mngr.getConnection(), getData('schema.xml'),
getData('add_response.txt'))
proc.index(self.foo)
mngr.closeConnection()
self.assertEqual(len(output), 2)
self.failUnless(output.get().startswith(self.schema_request))
self.assertEqual(sortFields(output.get()), getData('add_request.txt'))
def testUnindexingWithUniqueKeyMissing(self):
fakehttp(self.conn, getData('simple_schema.xml')) # fake schema response
self.mngr.getSchema() # read and cache the schema
response = getData('delete_response.txt')
output = fakehttp(self.conn, response) # fake delete response
foo = Foo(id='500', name='foo')
self.proc.unindex(foo) # unindexing sends data
self.assertEqual(len(output), 0) # nothing happened...
self.assertEqual(self.log, [
'schema is missing unique key, skipping unindexing of %r', foo])
def testIndexObject(self):
output = []
connection = self.proc.getConnection()
responses = getData('add_response.txt'), getData('commit_response.txt')
output = fakehttp(connection, *responses) # fake responses
self.folder.processForm(values={'title': 'Foo'}) # updating sends
self.assertEqual(self.folder.Title(), 'Foo')
self.assertEqual(str(output), '', 'reindexed unqueued!')
commit() # indexing happens on commit
required = '<field name="Title">Foo</field>'
self.assert_(str(output).find(required) > 0, '"title" data not found')
def test_add_with_boost_values(self):
add_request = getData('add_request_with_boost_values.txt')
add_response = getData('add_response.txt')
c = SolrConnection(host='localhost:8983', persistent=True)
output = fakehttp(c, add_response)
boost = {'': 2, 'id': 0.5, 'name': 5}
c.add(boost_values=boost, id='500', name='python test doc')
res = c.flush()
self.assertEqual(len(res), 1) # one request was sent
res = res[0]
self.failUnlessEqual(str(output), add_request)
def test_add_with_boost_values(self):
add_request = getData('add_request_with_boost_values.txt')
add_response = getData('add_response.txt')
c = SolrConnection(host='localhost:8983', persistent=True)
output = fakehttp(c, add_response)
boost = {'': 2, 'id': 0.5, 'name': 5}
c.add(boost_values=boost, id='500', name='python test doc')
res = c.flush()
self.assertEqual(len(res), 1) # one request was sent
res = res[0]
self.failUnlessEqual(str(output), add_request)
def test_search(self):
search_request = getData('search_request.txt')
search_response = getData('search_response.txt')
c = SolrConnection(host='localhost:8983', persistent=True)
output = fakehttp(c, search_response)
res = c.search(
q='+id:[* TO *]', fl='* score', wt='xml', rows='10', indent='on')
res = fromstring(res.read())
normalize = lambda x: sorted(x.split('&')) # sort request params
self.assertEqual(normalize(output.get()), normalize(search_request))
self.failUnless(res.find(('.//doc')))
def testIndexObject(self):
output = []
connection = self.proc.getConnection()
responses = getData('add_response.txt'), getData('commit_response.txt')
output = fakehttp(connection, *responses) # fake responses
self.folder.processForm(values={'title': 'Foo'}) # updating sends
self.assertEqual(self.folder.Title(), 'Foo')
self.assertEqual(str(output), '', 'reindexed unqueued!')
commit() # indexing happens on commit
required = '<field name="Title">Foo</field>'
self.assert_(str(output).find(required) > 0, '"title" data not found')
def testLocalConnections(self):
config = getConfig()
config.atomic_updates = True
mngr = SolrConnectionManager(active=True)
proc = SolrIndexProcessor(mngr)
mngr.setHost(active=True)
schema = getData("schema.xml")
log = []
def runner():
# fake schema response on solr connection - caches the schema
fakehttp(mngr.getConnection(), getData("schema.xml"))
mngr.getConnection().get_schema()
fakehttp(mngr.getConnection(), schema) # fake schema response
# read and cache the schema
mngr.getSchema()
response = getData("add_response.txt")
# fake add response
output = fakehttp(mngr.getConnection(), response)
# indexing sends data
proc.index(Foo(id="500", name="python test doc"))
mngr.closeConnection()
log.append(str(output))
log.append(proc)
log.append(mngr.getConnection())
# after the runner was set up, another thread can be created and
# started; its output should contain the proper indexing request,
# whereas the main thread's connection remain idle; the latter
# cannot be checked directly, but the connection object would raise
# an exception if it was used to send a request without setting up
# a fake response beforehand...
thread = Thread(target=runner)
thread.start()
thread.join()
conn = mngr.getConnection() # get this thread's connection
fakehttp(conn, schema) # fake schema response
mngr.getSchema() # read and cache the schema
mngr.closeConnection()
mngr.setHost(active=False)
self.assertEqual(len(log), 3)
self.assertEqual(
sortFields(log[0].encode("utf-8")), getData("add_request.txt").rstrip(b"\n")
)
self.failUnless(isinstance(log[1], SolrIndexProcessor))
self.failUnless(isinstance(log[2], SolrConnection))
self.failUnless(isinstance(proc, SolrIndexProcessor))
self.failUnless(isinstance(conn, SolrConnection))
self.assertEqual(log[1], proc) # processors should be the same...
self.assertNotEqual(log[2], conn) # but not the connections
def setUp(self):
replies = (getData('plone_schema.xml'), getData('add_response.txt'),
getData('add_response.txt'), getData('add_response.txt'),
getData('commit_response.txt'))
self.proc = queryUtility(ISolrConnectionManager)
self.proc.setHost(active=True)
conn = self.proc.getConnection()
fakehttp(conn, *replies) # fake schema response
self.proc.getSchema() # read and cache the schema
self.portal = self.layer['portal']
setRoles(self.portal, TEST_USER_ID, ['Manager'])
self.portal.invokeFactory('Folder', id='folder')
self.folder = self.portal.folder
commit()
def testLocalConnections(self):
config = getConfig()
config.atomic_updates = True
mngr = SolrConnectionManager(active=True)
proc = SolrIndexProcessor(mngr)
mngr.setHost(active=True)
schema = getData('schema.xml')
log = []
def runner():
# fake schema response on solr connection - caches the schema
fakehttp(mngr.getConnection(), getData('schema.xml'))
mngr.getConnection().get_schema()
fakehttp(mngr.getConnection(), schema) # fake schema response
# read and cache the schema
mngr.getSchema()
response = getData('add_response.txt')
# fake add response
output = fakehttp(mngr.getConnection(), response)
# indexing sends data
proc.index(Foo(id='500', name='python test doc'))
mngr.closeConnection()
log.append(str(output))
log.append(proc)
log.append(mngr.getConnection())
# after the runner was set up, another thread can be created and
# started; its output should contain the proper indexing request,
# whereas the main thread's connection remain idle; the latter
# cannot be checked directly, but the connection object would raise
# an exception if it was used to send a request without setting up
# a fake response beforehand...
thread = Thread(target=runner)
thread.start()
thread.join()
conn = mngr.getConnection() # get this thread's connection
fakehttp(conn, schema) # fake schema response
mngr.getSchema() # read and cache the schema
mngr.closeConnection()
mngr.setHost(active=False)
self.assertEqual(len(log), 3)
self.assertEqual(sortFields(log[0]), getData(
'add_request.txt').rstrip('\n'))
self.failUnless(isinstance(log[1], SolrIndexProcessor))
self.failUnless(isinstance(log[2], SolrConnection))
self.failUnless(isinstance(proc, SolrIndexProcessor))
self.failUnless(isinstance(conn, SolrConnection))
self.assertEqual(log[1], proc) # processors should be the same...
self.assertNotEqual(log[2], conn) # but not the connections
def testIndexAccessorRaises(self):
response = getData("add_response.txt")
# fake add response
output = fakehttp(self.mngr.getConnection(), response)
def brokenfunc():
raise ValueError
self.proc.index(
Foo(id="500", name="python test doc", text=brokenfunc)
) # indexing sends data
self.assertEqual(
sortFields(str(output).encode("utf-8")),
getData("add_request.txt").rstrip(b"\n"),
)
def testIndexObject(self):
output = []
connection = self.proc.getConnection()
connection.get_schema() # cache schema to avoid multiple calls
responses = (getData('plone_schema.xml'),
getData('commit_response.txt'))
output = fakehttp(connection, *responses) # fake responses
self.folder.title = 'Foo'
self.portal.invokeFactory('Folder', id='myfolder', title='Foo')
self.assertEqual(str(output), '', 'reindexed unqueued!')
commit() # indexing happens on commit
required = '<field name="Title" update="set">Foo</field>'
self.assertTrue(str(output).find(required) > 0,
'"title" data not found')
def testIndexObject(self):
output = []
connection = self.proc.getConnection()
connection.get_schema() # cache schema to avoid multiple calls
responses = (getData('plone_schema.xml'),
getData('commit_response.txt'))
output = fakehttp(connection, *responses) # fake responses
self.folder.title = 'Foo'
self.portal.invokeFactory('Folder', id='myfolder', title='Foo')
self.assertEqual(str(output), '', 'reindexed unqueued!')
commit() # indexing happens on commit
required = '<field name="Title" update="set">Foo</field>'
self.assertTrue(str(output).find(required) > 0,
'"title" data not found')
def testIndexingWithUniqueKeyMissing(self):
# fake schema response
fakehttp(self.conn, getData("simple_schema.xml"))
# read and cache the schema
self.mngr.getSchema()
response = getData("add_response.txt")
output = fakehttp(self.conn, response) # fake add response
foo = Foo(id="500", name="foo")
# indexing sends data
self.proc.index(foo)
# nothing happened...
self.assertEqual(len(output), 0)
self.assertEqual(
self.log, ["schema is missing unique key, skipping indexing of %r", foo]
)
def setUp(self):
replies = (getData('plone_schema.xml'), getData('add_response.txt'),
getData('add_response.txt'),
getData('add_response.txt'),
getData('commit_response.txt'))
self.proc = queryUtility(ISolrConnectionManager)
self.proc.setHost(active=True)
conn = self.proc.getConnection()
fakehttp(conn, *replies) # fake schema response
self.proc.getSchema() # read and cache the schema
self.portal = self.layer['portal']
setRoles(self.portal, TEST_USER_ID, ['Manager'])
self.portal.invokeFactory('Folder', id='folder')
self.folder = self.portal.folder
commit()
def testParseDateFacetSearchResults(self):
facet_xml_response = getData('date_facet_xml_response.txt')
response = SolrResponse(facet_xml_response)
results = response.response # the result set is named 'response'
self.assertEqual(results.numFound, '42')
self.assertEqual(results.start, '0')
self.assertEqual(len(results), 0)
headers = response.responseHeader
self.assertEqual(type(headers), type({}))
self.assertEqual(headers['status'], 0)
self.assertEqual(headers['QTime'], 5)
self.assertEqual(headers['params']['facet.date'], 'timestamp')
self.assertEqual(headers['params']['facet.date.start'],
'NOW/DAY-5DAYS')
self.assertEqual(headers['params']['facet.date.end'], 'NOW/DAY+1DAY')
self.assertEqual(headers['params']['facet.date.gap'], '+1DAY')
self.assertEqual(headers['params']['rows'], '0')
self.assertEqual(headers['params']['facet'], 'true')
self.assertEqual(headers['params']['indent'], 'true')
self.assertEqual(headers['params']['q'], '*:*')
counts = response.facet_counts
self.assertEqual(type(counts), type({}))
self.assertEqual(counts['facet_queries'], {})
self.assertEqual(counts['facet_fields'], {})
timestamps = counts['facet_dates']['timestamp']
self.assertEqual(timestamps['2007-08-11T00:00:00.000Z'], 1)
self.assertEqual(timestamps['2007-08-12T00:00:00.000Z'], 5)
self.assertEqual(timestamps['2007-08-13T00:00:00.000Z'], 3)
self.assertEqual(timestamps['2007-08-14T00:00:00.000Z'], 7)
self.assertEqual(timestamps['2007-08-15T00:00:00.000Z'], 2)
self.assertEqual(timestamps['2007-08-16T00:00:00.000Z'], 16)
self.assertEqual(timestamps['gap'], '+1DAY')
self.assertEqual(timestamps['end'], DateTime('2007-08-17 GMT'))
def testParseFacetSearchResults(self):
facet_xml_response = getData('facet_xml_response.txt')
response = SolrResponse(facet_xml_response)
results = response.response # the result set is named 'response'
self.assertEqual(results.numFound, '1')
self.assertEqual(results.start, '0')
self.assertEqual(len(results), 0)
headers = response.responseHeader
self.assertEqual(type(headers), type({}))
self.assertEqual(headers['status'], 0)
self.assertEqual(headers['QTime'], 1)
self.assertEqual(headers['params']['facet.limit'], '-1')
self.assertEqual(headers['params']['rows'], '0')
self.assertEqual(headers['params']['facet'], 'true')
self.assertEqual(headers['params']['facet.field'], ['cat', 'inStock'])
self.assertEqual(headers['params']['indent'], '10')
self.assertEqual(headers['params']['q'], 'solr')
counts = response.facet_counts
self.assertEqual(type(counts), type({}))
self.assertEqual(counts['facet_queries'], {})
self.assertEqual(counts['facet_fields']['cat']['electronics'], 0)
self.assertEqual(counts['facet_fields']['cat']['monitor'], 0)
self.assertEqual(counts['facet_fields']['cat']['search'], 1)
self.assertEqual(counts['facet_fields']['cat']['software'], 1)
self.assertEqual(counts['facet_fields']['inStock']['true'], 1)
def testParseDateFacetSearchResults(self):
facet_xml_response = getData("date_facet_xml_response.txt")
response = SolrResponse(facet_xml_response)
results = response.response # the result set is named 'response'
self.assertEqual(results.numFound, "42")
self.assertEqual(results.start, "0")
self.assertEqual(len(results), 0)
headers = response.responseHeader
self.assertEqual(type(headers), type({}))
self.assertEqual(headers["status"], 0)
self.assertEqual(headers["QTime"], 5)
self.assertEqual(headers["params"]["facet.date"], "timestamp")
self.assertEqual(headers["params"]["facet.date.start"], "NOW/DAY-5DAYS")
self.assertEqual(headers["params"]["facet.date.end"], "NOW/DAY+1DAY")
self.assertEqual(headers["params"]["facet.date.gap"], "+1DAY")
self.assertEqual(headers["params"]["rows"], "0")
self.assertEqual(headers["params"]["facet"], "true")
self.assertEqual(headers["params"]["indent"], "true")
self.assertEqual(headers["params"]["q"], "*:*")
counts = response.facet_counts
self.assertEqual(type(counts), type({}))
self.assertEqual(counts["facet_queries"], {})
self.assertEqual(counts["facet_fields"], {})
timestamps = counts["facet_dates"]["timestamp"]
self.assertEqual(timestamps["2007-08-11T00:00:00.000Z"], 1)
self.assertEqual(timestamps["2007-08-12T00:00:00.000Z"], 5)
self.assertEqual(timestamps["2007-08-13T00:00:00.000Z"], 3)
self.assertEqual(timestamps["2007-08-14T00:00:00.000Z"], 7)
self.assertEqual(timestamps["2007-08-15T00:00:00.000Z"], 2)
self.assertEqual(timestamps["2007-08-16T00:00:00.000Z"], 16)
self.assertEqual(timestamps["gap"], "+1DAY")
self.assertEqual(timestamps["end"].ISO8601(), DateTime("2007-08-17 GMT").ISO8601())
def testParseFacetSearchResults(self):
facet_xml_response = getData("facet_xml_response.txt")
response = SolrResponse(facet_xml_response)
results = response.response # the result set is named 'response'
self.assertEqual(results.numFound, "1")
self.assertEqual(results.start, "0")
self.assertEqual(len(results), 0)
headers = response.responseHeader
self.assertEqual(type(headers), type({}))
self.assertEqual(headers["status"], 0)
self.assertEqual(headers["QTime"], 1)
self.assertEqual(headers["params"]["facet.limit"], "-1")
self.assertEqual(headers["params"]["rows"], "0")
self.assertEqual(headers["params"]["facet"], "true")
self.assertEqual(headers["params"]["facet.field"], ["cat", "inStock"])
self.assertEqual(headers["params"]["indent"], "10")
self.assertEqual(headers["params"]["q"], "solr")
counts = response.facet_counts
self.assertEqual(type(counts), type({}))
self.assertEqual(counts["facet_queries"], {})
self.assertEqual(counts["facet_fields"]["cat"]["electronics"], 0)
self.assertEqual(counts["facet_fields"]["cat"]["monitor"], 0)
self.assertEqual(counts["facet_fields"]["cat"]["search"], 1)
self.assertEqual(counts["facet_fields"]["cat"]["software"], 1)
self.assertEqual(counts["facet_fields"]["inStock"]["true"], 1)
def testParseComplexSearchResults(self):
complex_xml_response = getData("complex_xml_response.txt")
response = SolrResponse(complex_xml_response)
results = response.response # the result set is named 'response'
self.assertEqual(results.numFound, "2")
self.assertEqual(results.start, "0")
self.assertEqual(len(results), 2)
first = results[0]
self.assertEqual(first.cat, ["software", "search"])
self.assertEqual(len(first.features), 7)
self.assertEqual([type(x).__name__ for x in first.features], ["str"] * 6 + ["unicode"])
self.assertEqual(first.id, "SOLR1000")
self.assertEqual(first.inStock, True)
self.assertEqual(first.incubationdate_dt.ISO8601(), DateTime("2006/01/17 GMT").ISO8601())
self.assertEqual(first.manu, "Apache Software Foundation")
self.assertEqual(first.popularity, 10)
self.assertEqual(first.price, 0.0)
headers = response.responseHeader
self.assertEqual(headers["status"], 0)
self.assertEqual(headers["QTime"], 0)
self.assertEqual(headers["params"]["indent"], "on")
self.assertEqual(headers["params"]["rows"], "10")
self.assertEqual(headers["params"]["start"], "0")
self.assertEqual(headers["params"]["q"], "id:[* TO *]")
self.assertEqual(headers["params"]["version"], "2.2")
def testParseConfig(self):
schema_xml = getData("schema.xml")
schema = SolrSchema(schema_xml.split("\n\n", 1)[1])
self.assertEqual(len(schema), 21) # 21 items defined in schema.xml
self.assertEqual(schema["defaultSearchField"], "text")
self.assertEqual(schema["uniqueKey"], "id")
self.assertEqual(schema["solrQueryParser"].defaultOperator, "OR")
self.assertEqual(schema["requiredFields"], ["id", "name"])
self.assertEqual(schema["id"].type, "string")
self.assertEqual(schema["id"].class_, "solr.StrField")
self.assertEqual(schema["id"].required, True)
self.assertEqual(schema["id"].omitNorms, True)
self.assertEqual(schema["id"].multiValued, False)
self.assertEqual(schema["cat"].class_, "solr.TextField")
self.assertEqual(schema["cat"].required, False)
self.assertEqual(schema["cat"].multiValued, True)
self.assertEqual(schema["cat"].termVectors, True)
self.assertEqual(schema["sku"].positionIncrementGap, "100")
self.assertEqual(schema.features.multiValued, True)
self.assertEqual(schema.weight.class_, "solr.SortableFloatField")
self.assertEqual(schema.popularity.class_, "solr.SortableIntField")
self.assertEqual(schema.inStock.class_, "solr.BoolField")
self.assertEqual(schema.timestamp.class_, "solr.DateField")
self.assertEqual(schema.timestamp.default, "NOW")
self.assertEqual(schema.timestamp.multiValued, False)
self.assertEqual(schema.timestamp.indexed, True)
self.assertEqual(schema.word.indexed, False)
fields = schema.values()
self.assertEqual(len([f for f in fields if getattr(f, "required", False)]), 2)
self.assertEqual(len([f for f in fields if getattr(f, "multiValued", False)]), 3)
def testExtraRequest(self):
# basically the same as `testThreeRequests`, except it
# tests adding fake responses consecutively
mngr = SolrConnectionManager(active=True)
proc = SolrIndexProcessor(mngr)
conn = mngr.getConnection()
output = fakehttp(conn, getData('schema.xml'))
fakemore(conn, getData('add_response.txt'))
proc.index(self.foo)
fakemore(conn, getData('delete_response.txt'))
proc.unindex(self.foo)
mngr.closeConnection()
self.assertEqual(len(output), 3)
self.failUnless(output.get().startswith(self.schema_request))
self.assertEqual(sortFields(output.get()), getData('add_request.txt'))
self.assertEqual(output.get(), getData('delete_request.txt'))
def testNoIndexingWithoutAllRequiredFields(self):
response = getData('dummy_response.txt')
# fake add response
output = fakehttp(self.mngr.getConnection(), response)
# indexing sends data
self.proc.index(Foo(id='500'))
self.assertEqual(str(output), '')
def testParseComplexSearchResults(self):
complex_xml_response = getData('complex_xml_response.txt')
response = SolrResponse(complex_xml_response)
results = response.response # the result set is named 'response'
self.assertEqual(results.numFound, '2')
self.assertEqual(results.start, '0')
self.assertEqual(len(results), 2)
first = results[0]
self.assertEqual(first.cat, ['software', 'search'])
self.assertEqual(len(first.features), 7)
self.assertEqual([type(x).__name__ for x in first.features],
['str'] * 6 + ['unicode'])
self.assertEqual(first.id, 'SOLR1000')
self.assertEqual(first.inStock, True)
self.assertEqual(first.incubationdate_dt, DateTime('2006/01/17 GMT'))
self.assertEqual(first.manu, 'Apache Software Foundation')
self.assertEqual(first.popularity, 10)
self.assertEqual(first.price, 0.0)
headers = response.responseHeader
self.assertEqual(headers['status'], 0)
self.assertEqual(headers['QTime'], 0)
self.assertEqual(headers['params']['indent'], 'on')
self.assertEqual(headers['params']['rows'], '10')
self.assertEqual(headers['params']['start'], '0')
self.assertEqual(headers['params']['q'], 'id:[* TO *]')
self.assertEqual(headers['params']['version'], '2.2')
def testParseConfig(self):
schema_xml = getData('schema.xml')
schema = SolrSchema(schema_xml.split('\n\n', 1)[1])
self.assertEqual(len(schema), 21) # 21 items defined in schema.xml
self.assertEqual(schema['defaultSearchField'], 'text')
self.assertEqual(schema['uniqueKey'], 'id')
self.assertEqual(schema['solrQueryParser'].defaultOperator, 'OR')
self.assertEqual(schema['requiredFields'], ['id', 'name'])
self.assertEqual(schema['id'].type, 'string')
self.assertEqual(schema['id'].class_, 'solr.StrField')
self.assertEqual(schema['id'].required, True)
self.assertEqual(schema['id'].omitNorms, True)
self.assertEqual(schema['id'].multiValued, False)
self.assertEqual(schema['cat'].class_, 'solr.TextField')
self.assertEqual(schema['cat'].required, False)
self.assertEqual(schema['cat'].multiValued, True)
self.assertEqual(schema['cat'].termVectors, True)
self.assertEqual(schema['sku'].positionIncrementGap, '100')
self.assertEqual(schema.features.multiValued, True)
self.assertEqual(schema.weight.class_, 'solr.SortableFloatField')
self.assertEqual(schema.popularity.class_, 'solr.SortableIntField')
self.assertEqual(schema.inStock.class_, 'solr.BoolField')
self.assertEqual(schema.timestamp.class_, 'solr.DateField')
self.assertEqual(schema.timestamp.default, 'NOW')
self.assertEqual(schema.timestamp.multiValued, False)
self.assertEqual(schema.timestamp.indexed, True)
self.assertEqual(schema.word.indexed, False)
fields = schema.values()
self.assertEqual(len([f for f in fields if
getattr(f, 'required', False)]), 2)
self.assertEqual(len([f for f in fields if
getattr(f, 'multiValued', False)]), 3)
def testDateIndexingWithPythonDateTime(self):
foo = Foo(id='gerken', name='patrick', cat='nerd', timestamp=datetime(1980, 9, 29, 14, 02))
response = getData('add_response.txt')
output = fakehttp(self.mngr.getConnection(), response) # fake add response
self.proc.index(foo)
required = '<field name="timestamp">1980-09-29T14:02:00.000Z</field>'
self.assert_(str(output).find(required) > 0, '"date" data not found')