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_read_bulks_and_assume_metadata_line_in_source_file(self):
data = [
'{"index": {"_index": "test_index", "_type": "test_type"}',
'{"key": "value1"}',
'{"index": {"_index": "test_index", "_type": "test_type"}',
'{"key": "value2"}',
'{"index": {"_index": "test_index", "_type": "test_type"}',
'{"key": "value3"}',
'{"index": {"_index": "test_index", "_type": "test_type"}',
'{"key": "value4"}',
'{"index": {"_index": "test_index", "_type": "test_type"}',
'{"key": "value5"}',
'{"index": {"_index": "test_index", "_type": "test_type"}',
'{"key": "value6"}',
'{"index": {"_index": "test_index", "_type": "test_type"}',
'{"key": "value7"}'
]
bulk_size = 3
source = params.Slice(io.StringAsFileSource, 0, len(data))
am_handler = params.SourceActionMetaData(source)
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, 3, 1]
# lines should include meta-data
expected_line_sizes = [6, 6, 2]
self.assert_bulks_sized(reader, expected_bulk_sizes,
expected_line_sizes)
def test_read_bulks_and_assume_metadata_line_in_source_file(self):
data = [
'{"index": {"_index": "test_index", "_type": "test_type"}',
'{"key": "value1"}',
'{"index": {"_index": "test_index", "_type": "test_type"}',
'{"key": "value2"}',
'{"index": {"_index": "test_index", "_type": "test_type"}',
'{"key": "value3"}',
'{"index": {"_index": "test_index", "_type": "test_type"}',
'{"key": "value4"}',
'{"index": {"_index": "test_index", "_type": "test_type"}',
'{"key": "value5"}',
'{"index": {"_index": "test_index", "_type": "test_type"}',
'{"key": "value6"}',
'{"index": {"_index": "test_index", "_type": "test_type"}',
'{"key": "value7"}'
]
bulk_size = 3
source = params.Slice(StringAsFileSource, 0, len(data))
am_handler = params.SourceActionMetaData(source)
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, 6, 2]
self.assert_bulks_sized(reader, expected_bulk_sizes)
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 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
reader = params.IndexDataReader(data,
docs_to_index=5,
conflicting_ids=None,
index_name="test_index",
type_name="test_type",
bulk_size=bulk_size,
file_source=StringAsFileSource)
# always double the amount as one line contains the data and one line contains the index command
expected_bulk_lengths = [6, 4]
with reader:
bulk_index = 0
for bulk in reader:
self.assertEqual(expected_bulk_lengths[bulk_index], len(bulk))
bulk_index += 1
def test_read_bulk_larger_than_number_of_docs(self):
data = [
'{"key": "value1"}', '{"key": "value2"}', '{"key": "value3"}',
'{"key": "value4"}', '{"key": "value5"}'
]
bulk_size = 50
reader = params.IndexDataReader(data,
docs_to_index=len(data),
conflicting_ids=None,
index_name="test_index",
type_name="test_type",
bulk_size=bulk_size,
file_source=StringAsFileSource)
with reader:
for bulk in reader:
self.assertEqual(len(data) * 2, len(bulk))
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_read_bulks_and_assume_no_metadata(self):
data = [
'{"key": "value1"}',
'{"key": "value2"}',
'{"key": "value3"}',
'{"key": "value4"}',
'{"key": "value5"}',
'{"key": "value6"}',
'{"key": "value7"}'
]
bulk_size = 3
source = params.Slice(io.StringAsFileSource, 0, len(data))
am_handler = params.NoneActionMetaData()
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")
# no meta-data, hence line numbers and bulk sizes need to be identical
expected_bulk_sizes = [3, 3, 1]
self.assert_bulks_sized(reader, expected_bulk_sizes, expected_bulk_sizes)
