def test_read_bulk_smaller_than_number_of_docs_and_multiple_clients(self):
data = [
'{"key": "value1"}',
'{"key": "value2"}',
'{"key": "value3"}',
'{"key": "value4"}',
'{"key": "value5"}',
'{"key": "value6"}',
'{"key": "value7"}',
]
bulk_size = 3
# only 5 documents to index for this client
source = params.Slice(io.StringAsFileSource, 0, 5)
am_handler = params.GenerateActionMetaData("test_index",
"test_type",
conflicting_ids=None)
reader = params.IndexDataReader(data,
batch_size=bulk_size,
bulk_size=bulk_size,
file_source=source,
action_metadata=am_handler,
index_name="test_index",
type_name="test_type")
expected_bulk_sizes = [3, 2]
# lines should include meta-data
expected_line_sizes = [6, 4]
self.assert_bulks_sized(reader, expected_bulk_sizes,
expected_line_sizes)
def test_generate_action_meta_data_without_id_conflicts(self):
self.assertEqual(
'{"index": {"_index": "test_index", "_type": "test_type"}}',
next(
params.GenerateActionMetaData("test_index",
"test_type",
conflicting_ids=None)))
def test_read_bulk_with_offset(self):
data = [
'{"key": "value1"}', '{"key": "value2"}', '{"key": "value3"}',
'{"key": "value4"}', '{"key": "value5"}'
]
bulk_size = 50
source = params.Slice(io.StringAsFileSource, 3, len(data))
am_handler = params.GenerateActionMetaData("test_index",
"test_type",
conflicting_ids=None)
reader = params.IndexDataReader(data,
batch_size=bulk_size,
bulk_size=bulk_size,
file_source=source,
action_metadata=am_handler,
index_name="test_index",
type_name="test_type")
expected_bulk_sizes = [(len(data) - 3)]
# lines should include meta-data
expected_line_sizes = [(len(data) - 3) * 2]
self.assert_bulks_sized(reader, expected_bulk_sizes,
expected_line_sizes)
def test_read_bulk_smaller_than_number_of_docs_and_multiple_clients(self):
data = [
'{"key": "value1"}',
'{"key": "value2"}',
'{"key": "value3"}',
'{"key": "value4"}',
'{"key": "value5"}',
'{"key": "value6"}',
'{"key": "value7"}',
]
bulk_size = 3
# only 5 documents to index for this client
source = params.Slice(StringAsFileSource, 0, 5)
am_handler = params.GenerateActionMetaData("test_index",
"test_type",
conflicting_ids=None)
reader = params.IndexDataReader(data,
batch_size=bulk_size,
bulk_size=bulk_size,
file_source=source,
action_metadata=am_handler,
index_name="test_index",
type_name="test_type")
# always double the amount as one line contains the data and one line contains the index command
expected_bulk_sizes = [6, 4]
self.assert_bulks_sized(reader, expected_bulk_sizes)
def create_reader(bulk_size):
metadata = params.GenerateActionMetaData(index_name="test-idx",
type_name=None)
source = params.Slice(StaticSource, 0, sys.maxsize)
reader = params.MetadataIndexDataReader(data_file="bogus",
batch_size=bulk_size,
bulk_size=bulk_size,
file_source=source,
action_metadata=metadata,
index_name="test-idx",
type_name=None)
return reader
def test_read_bulk_larger_than_number_of_docs(self):
data = [
'{"key": "value1"}',
'{"key": "value2"}',
'{"key": "value3"}',
'{"key": "value4"}',
'{"key": "value5"}'
]
bulk_size = 50
source = params.Slice(io.StringAsFileSource, 0, len(data))
am_handler = params.GenerateActionMetaData("test_index", "test_type", conflicting_ids=None)
reader = params.IndexDataReader(data, batch_size=bulk_size, bulk_size=bulk_size, file_source=source, action_metadata=am_handler,
index_name="test_index", type_name="test_type")
expected_bulk_sizes = [len(data) * 2]
self.assert_bulks_sized(reader, expected_bulk_sizes)
def test_generate_action_meta_data_with_id_conflicts(self):
pseudo_random_sequence = iter([
# first column == 3 -> we'll draw a "random" id, second column == "random" id
3,
1,
3,
3,
3,
2,
0,
3,
0
])
generator = params.GenerateActionMetaData(
"test_index",
"test_type",
conflicting_ids=[100, 200, 300, 400],
rand=lambda x, y: next(pseudo_random_sequence))
# first one is always not drawn from a random index
self.assertEqual(
'{"index": {"_index": "test_index", "_type": "test_type", "_id": "100"}}',
next(generator))
# now we start using random ids, i.e. look in the first line of the pseudo-random sequence
self.assertEqual(
'{"index": {"_index": "test_index", "_type": "test_type", "_id": "200"}}',
next(generator))
self.assertEqual(
'{"index": {"_index": "test_index", "_type": "test_type", "_id": "400"}}',
next(generator))
self.assertEqual(
'{"index": {"_index": "test_index", "_type": "test_type", "_id": "300"}}',
next(generator))
# "random" returns 0 instead of 3 -> we draw the next sequential one, which is 200
self.assertEqual(
'{"index": {"_index": "test_index", "_type": "test_type", "_id": "200"}}',
next(generator))
# and we're back to random
self.assertEqual(
'{"index": {"_index": "test_index", "_type": "test_type", "_id": "100"}}',
next(generator))
