mirror of
https://gitee.com/milvus-io/milvus.git
synced 2026-01-07 19:31:51 +08:00
15ce8aedd8
related issue: #46616 <!-- This is an auto-generated comment: release notes by coderabbit.ai --> - Core invariant: these tests assume the v2 group-by search implementation (TestMilvusClientV2Base and pymilvus v2 APIs such as AnnSearchRequest/WeightedRanker) is functionally correct; the PR extends coverage to validate group-by semantics when using JSON fields and dynamic fields (see tests/python_client/milvus_client_v2/test_milvus_client_search_group_by.py — TestGroupSearch.setup_class and parametrized group_by_field cases). - Logic removed/simplified: legacy v1 test scaffolding and duplicated parametrized fixtures/test permutations were consolidated into v2-focused suites (TestGroupSearch now inherits TestMilvusClientV2Base; old TestGroupSearch/TestcaseBase patterns and large blocks in test_mix_scenes were removed) to avoid redundant fixture permutations and duplicate assertions while reusing shared helpers in common_func (e.g., gen_scalar_field, gen_row_data_by_schema) and common_type constants. - Why this does NOT introduce data loss or behavior regression: only test code, test helpers, and test imports were changed — no production/server code altered. Test helper changes are backward-compatible (gen_scalar_field forces primary key nullable=False and only affects test data generation paths in tests/python_client/common/common_func.py; get_field_dtype_by_field_name now accepts schema dicts/ORM schemas and is used only by tests to choose vector generation) and collection creation/insertion in tests use the same CollectionSchema/FieldSchema paths, so production storage/serialization logic is untouched. - New capability (test addition): adds v2 test coverage for group-by search over JSON and dynamic fields plus related scenarios — pagination, strict/non-strict group_size, min/max group constraints, multi-field group-bys and binary vector cases — implemented in tests/python_client/milvus_client_v2/test_milvus_client_search_group_by.py to address issue #46616. <!-- end of auto-generated comment: release notes by coderabbit.ai --> --------- Signed-off-by: yanliang567 <yanliang.qiao@zilliz.com>
1860 lines
110 KiB
Python
1860 lines
110 KiB
Python
import pytest
|
|
|
|
from base.client_v2_base import TestMilvusClientV2Base
|
|
from utils.util_log import test_log as log
|
|
from common import common_func as cf
|
|
from common import common_type as ct
|
|
from common.common_type import CaseLabel, CheckTasks
|
|
from utils.util_pymilvus import *
|
|
from pymilvus import DataType
|
|
|
|
prefix = "client_index"
|
|
epsilon = ct.epsilon
|
|
default_nb = ct.default_nb
|
|
default_nb_medium = ct.default_nb_medium
|
|
default_nq = ct.default_nq
|
|
default_dim = ct.default_dim
|
|
default_limit = ct.default_limit
|
|
default_search_exp = "id >= 0"
|
|
exp_res = "exp_res"
|
|
default_search_string_exp = "varchar >= \"0\""
|
|
default_search_mix_exp = "int64 >= 0 && varchar >= \"0\""
|
|
default_invaild_string_exp = "varchar >= 0"
|
|
default_json_search_exp = "json_field[\"number\"] >= 0"
|
|
perfix_expr = 'varchar like "0%"'
|
|
default_search_field = ct.default_float_vec_field_name
|
|
default_search_params = ct.default_search_params
|
|
default_primary_key_field_name = "id"
|
|
default_vector_field_name = "vector"
|
|
default_multiple_vector_field_name = "vector_new"
|
|
default_float_field_name = ct.default_float_field_name
|
|
default_bool_field_name = ct.default_bool_field_name
|
|
default_string_field_name = ct.default_string_field_name
|
|
default_int32_array_field_name = ct.default_int32_array_field_name
|
|
default_string_array_field_name = ct.default_string_array_field_name
|
|
|
|
|
|
class TestMilvusClientIndexInvalid(TestMilvusClientV2Base):
|
|
""" Test case of search interface """
|
|
|
|
@pytest.fixture(scope="function", params=[False, True])
|
|
def auto_id(self, request):
|
|
yield request.param
|
|
|
|
@pytest.fixture(scope="function", params=["COSINE", "L2"])
|
|
def metric_type(self, request):
|
|
yield request.param
|
|
|
|
"""
|
|
******************************************************************
|
|
# The following are invalid base cases
|
|
******************************************************************
|
|
"""
|
|
|
|
@pytest.mark.tags(CaseLabel.L1)
|
|
@pytest.mark.parametrize("name", ["12-s", "12 s", "(mn)", "中文", "%$#"])
|
|
def test_milvus_client_index_invalid_collection_name(self, name):
|
|
"""
|
|
target: test index abnormal case
|
|
method: create index on invalid collection name
|
|
expected: raise exception
|
|
"""
|
|
client = self._client()
|
|
collection_name = cf.gen_unique_str(prefix)
|
|
# 1. create collection
|
|
self.create_collection(client, collection_name, default_dim, consistency_level="Strong")
|
|
self.release_collection(client, collection_name)
|
|
self.drop_index(client, collection_name, "vector")
|
|
# 2. prepare index params
|
|
index_params = self.prepare_index_params(client)[0]
|
|
index_params.add_index(field_name="vector")
|
|
# 3. create index
|
|
error = {ct.err_code: 100, ct.err_msg: f"collection not found[database=default][collection={name}"}
|
|
self.create_index(client, name, index_params,
|
|
check_task=CheckTasks.err_res, check_items=error)
|
|
self.drop_collection(client, collection_name)
|
|
|
|
@pytest.mark.tags(CaseLabel.L1)
|
|
@pytest.mark.parametrize("name", ["a".join("a" for i in range(256))])
|
|
def test_milvus_client_index_collection_name_over_max_length(self, name):
|
|
"""
|
|
target: test index abnormal case
|
|
method: create index on collection name over max length
|
|
expected: raise exception
|
|
"""
|
|
client = self._client()
|
|
collection_name = cf.gen_unique_str(prefix)
|
|
# 1. create collection
|
|
self.create_collection(client, collection_name, default_dim, consistency_level="Strong")
|
|
self.release_collection(client, collection_name)
|
|
self.drop_index(client, collection_name, "vector")
|
|
# 2. prepare index params
|
|
index_params = self.prepare_index_params(client)[0]
|
|
index_params.add_index(field_name="vector")
|
|
# 3. create index
|
|
error = {ct.err_code: 100, ct.err_msg: f"collection not found[database=default][collection={name}]"}
|
|
self.create_index(client, name, index_params,
|
|
check_task=CheckTasks.err_res, check_items=error)
|
|
self.drop_collection(client, collection_name)
|
|
|
|
@pytest.mark.tags(CaseLabel.L1)
|
|
def test_milvus_client_index_not_exist_collection_name(self):
|
|
"""
|
|
target: test index abnormal case
|
|
method: create index on not exist collection name
|
|
expected: raise exception
|
|
"""
|
|
client = self._client()
|
|
collection_name = cf.gen_unique_str(prefix)
|
|
not_existed_collection_name = cf.gen_unique_str("not_existed_collection")
|
|
# 1. create collection
|
|
self.create_collection(client, collection_name, default_dim, consistency_level="Strong")
|
|
self.release_collection(client, collection_name)
|
|
self.drop_index(client, collection_name, "vector")
|
|
# 2. prepare index params
|
|
index_params = self.prepare_index_params(client)[0]
|
|
index_params.add_index(field_name="vector")
|
|
# 3. create index
|
|
error = {ct.err_code: 100,
|
|
ct.err_msg: f"collection not found[database=default][collection={not_existed_collection_name}]"}
|
|
self.create_index(client, not_existed_collection_name, index_params,
|
|
check_task=CheckTasks.err_res, check_items=error)
|
|
self.drop_collection(client, collection_name)
|
|
|
|
@pytest.mark.tags(CaseLabel.L1)
|
|
@pytest.mark.skip(reason="pymilvus issue 1885")
|
|
@pytest.mark.parametrize("index", ["12-s", "12 s", "(mn)", "中文", "%$#", "a".join("a" for i in range(256))])
|
|
def test_milvus_client_index_invalid_index_type(self, index):
|
|
"""
|
|
target: test index abnormal case
|
|
method: create index on invalid index type
|
|
expected: raise exception
|
|
"""
|
|
client = self._client()
|
|
collection_name = cf.gen_unique_str(prefix)
|
|
# 1. create collection
|
|
self.create_collection(client, collection_name, default_dim, consistency_level="Strong")
|
|
self.release_collection(client, collection_name)
|
|
self.drop_index(client, collection_name, "vector")
|
|
# 2. prepare index params
|
|
index_params = self.prepare_index_params(client)[0]
|
|
index_params.add_index(field_name="vector", index_type=index)
|
|
# 3. create index
|
|
error = {ct.err_code: 100, ct.err_msg: f"can't find collection collection not "
|
|
f"found[database=default][collection=not_existed]"}
|
|
self.create_index(client, collection_name, index_params,
|
|
check_task=CheckTasks.err_res, check_items=error)
|
|
self.drop_collection(client, collection_name)
|
|
|
|
@pytest.mark.tags(CaseLabel.L1)
|
|
@pytest.mark.skip(reason="pymilvus issue 1885")
|
|
@pytest.mark.parametrize("metric", ["12-s", "12 s", "(mn)", "中文", "%$#", "a".join("a" for i in range(256))])
|
|
def test_milvus_client_index_invalid_metric_type(self, metric):
|
|
"""
|
|
target: test index abnormal case
|
|
method: create index on invalid metric type
|
|
expected: raise exception
|
|
"""
|
|
client = self._client()
|
|
collection_name = cf.gen_unique_str(prefix)
|
|
# 1. create collection
|
|
self.create_collection(client, collection_name, default_dim, consistency_level="Strong")
|
|
self.release_collection(client, collection_name)
|
|
self.drop_index(client, collection_name, "vector")
|
|
# 2. prepare index params
|
|
index_params = self.prepare_index_params(client)[0]
|
|
index_params.add_index(field_name="vector", metric_type=metric)
|
|
# 3. create index
|
|
error = {ct.err_code: 100, ct.err_msg: f"can't find collection collection not "
|
|
f"found[database=default][collection=not_existed]"}
|
|
self.create_index(client, collection_name, index_params,
|
|
check_task=CheckTasks.err_res, check_items=error)
|
|
self.drop_collection(client, collection_name)
|
|
|
|
@pytest.mark.tags(CaseLabel.L1)
|
|
def test_milvus_client_index_drop_index_before_release(self):
|
|
"""
|
|
target: test index abnormal case
|
|
method: drop index before release
|
|
expected: raise exception
|
|
"""
|
|
client = self._client()
|
|
collection_name = cf.gen_unique_str(prefix)
|
|
# 1. create collection
|
|
self.create_collection(client, collection_name, default_dim, consistency_level="Strong")
|
|
error = {ct.err_code: 1100, ct.err_msg: f"vector index cannot be dropped on loaded collection"}
|
|
self.drop_index(client, collection_name, "vector",
|
|
check_task=CheckTasks.err_res, check_items=error)
|
|
self.drop_collection(client, collection_name)
|
|
|
|
@pytest.mark.tags(CaseLabel.L1)
|
|
def test_milvus_client_create_multiple_diff_index_without_release(self):
|
|
"""
|
|
target: test index abnormal case
|
|
method: create different index on one field without release
|
|
expected: raise exception
|
|
"""
|
|
client = self._client()
|
|
collection_name = cf.gen_unique_str(prefix)
|
|
# 1. create collection
|
|
self.create_collection(client, collection_name, default_dim, consistency_level="Strong")
|
|
# 2. prepare index params
|
|
index_params = self.prepare_index_params(client)[0]
|
|
index_params.add_index(field_name="vector", index_type="IVF_FLAT", metric_type="L2")
|
|
# 3. create another index
|
|
error = {ct.err_code: 65535, ct.err_msg: "CreateIndex failed: at most one distinct index is allowed per field"}
|
|
self.create_index(client, collection_name, index_params,
|
|
check_task=CheckTasks.err_res, check_items=error)
|
|
self.drop_collection(client, collection_name)
|
|
|
|
@pytest.mark.tags(CaseLabel.L1)
|
|
@pytest.mark.parametrize("not_supported_index", ct.all_index_types[:-2])
|
|
def test_milvus_client_int8_vector_create_not_supported_cpu_index(self, not_supported_index):
|
|
"""
|
|
target: test create non-supported index on int8 vector
|
|
method: create non-supported index on int8 vector
|
|
expected: raise exception
|
|
"""
|
|
if not_supported_index in ct.int8_vector_index:
|
|
pytest.skip("This index is supported by int8 vector")
|
|
client = self._client()
|
|
collection_name = cf.gen_unique_str(prefix)
|
|
dim = 128
|
|
# 1. create collection
|
|
schema = self.create_schema(client, enable_dynamic_field=False)[0]
|
|
schema.add_field("id_string", DataType.VARCHAR, max_length=64, is_primary=True, auto_id=False)
|
|
schema.add_field("embeddings", DataType.INT8_VECTOR, dim=dim)
|
|
index_params = self.prepare_index_params(client)[0]
|
|
index_params.add_index("embeddings", metric_type="COSINE")
|
|
# 2. index_params.add_index("title")
|
|
self.create_collection(client, collection_name, dimension=dim, schema=schema, index_params=index_params)
|
|
self.release_collection(client, collection_name)
|
|
self.drop_index(client, collection_name, "embeddings")
|
|
# 3. prepare index params
|
|
index_params = self.prepare_index_params(client)[0]
|
|
index_params.add_index(field_name="embeddings", index_type=not_supported_index, metric_type="L2")
|
|
# 4. create another index
|
|
error = {ct.err_code: 1100,
|
|
ct.err_msg: f"data type Int8Vector can't build with this index {not_supported_index}: "
|
|
f"invalid parameter[expected=valid index params][actual=invalid index params]"}
|
|
self.create_index(client, collection_name, index_params,
|
|
check_task=CheckTasks.err_res, check_items=error)
|
|
self.drop_collection(client, collection_name)
|
|
|
|
@pytest.mark.tags(CaseLabel.L1)
|
|
@pytest.mark.parametrize("not_supported_index", ct.all_index_types[-2:])
|
|
def test_milvus_client_int8_vector_create_not_supported_GPU_index(self, not_supported_index):
|
|
"""
|
|
target: test create non-supported index on int8 vector
|
|
method: create non-supported index on int8 vector
|
|
expected: raise exception
|
|
"""
|
|
if not_supported_index in ct.int8_vector_index:
|
|
pytest.skip("This index is supported by int8 vector")
|
|
client = self._client()
|
|
collection_name = cf.gen_unique_str(prefix)
|
|
dim = 128
|
|
# 1. create collection
|
|
schema = self.create_schema(client, enable_dynamic_field=False)[0]
|
|
schema.add_field("id_string", DataType.VARCHAR, max_length=64, is_primary=True, auto_id=False)
|
|
schema.add_field("embeddings", DataType.INT8_VECTOR, dim=dim)
|
|
index_params = self.prepare_index_params(client)[0]
|
|
index_params.add_index("embeddings", metric_type="COSINE")
|
|
# 2. index_params.add_index("title")
|
|
self.create_collection(client, collection_name, dimension=dim, schema=schema, index_params=index_params)
|
|
self.release_collection(client, collection_name)
|
|
self.drop_index(client, collection_name, "embeddings")
|
|
# 3. prepare index params
|
|
index_params = self.prepare_index_params(client)[0]
|
|
index_params.add_index(field_name="embeddings", index_type=not_supported_index, metric_type="L2")
|
|
# 4. create another index
|
|
error = {ct.err_code: 1100, ct.err_msg: f"invalid parameter[expected=valid index][actual=invalid "
|
|
f"index type: {not_supported_index}"}
|
|
self.create_index(client, collection_name, index_params,
|
|
check_task=CheckTasks.err_res, check_items=error)
|
|
self.drop_collection(client, collection_name)
|
|
|
|
|
|
class TestMilvusClientIndexValid(TestMilvusClientV2Base):
|
|
""" Test case of index interface """
|
|
|
|
@pytest.fixture(scope="function", params=["COSINE", "L2", "IP"])
|
|
def metric_type(self, request):
|
|
yield request.param
|
|
|
|
"""
|
|
******************************************************************
|
|
# The following are valid base cases
|
|
******************************************************************
|
|
"""
|
|
|
|
@pytest.mark.tags(CaseLabel.L0)
|
|
@pytest.mark.parametrize("index", ct.all_index_types[:8])
|
|
def test_milvus_client_index_with_params(self, index):
|
|
"""
|
|
target: test index with user defined params
|
|
method: create connection, collection, index, insert and search
|
|
expected: index/search/query successfully
|
|
"""
|
|
metric_type = "L2"
|
|
client = self._client()
|
|
collection_name = cf.gen_unique_str(prefix)
|
|
# 1. create collection
|
|
self.create_collection(client, collection_name, default_dim, consistency_level="Strong")
|
|
self.release_collection(client, collection_name)
|
|
self.drop_index(client, collection_name, "vector")
|
|
res = self.list_indexes(client, collection_name)[0]
|
|
assert res == []
|
|
# 2. prepare index params
|
|
index_params = self.prepare_index_params(client)[0]
|
|
params = cf.get_index_params_params(index_type=index)
|
|
index_params.add_index(field_name="vector", index_type=index, params=params, metric_type=metric_type)
|
|
# 3. create index
|
|
self.create_index(client, collection_name, index_params)
|
|
# 4. insert
|
|
rng = np.random.default_rng(seed=19530)
|
|
rows = [{default_primary_key_field_name: i, default_vector_field_name: list(rng.random((1, default_dim))[0]),
|
|
default_float_field_name: i * 1.0, default_string_field_name: str(i)} for i in range(default_nb)]
|
|
self.insert(client, collection_name, rows)
|
|
# 5. load collection
|
|
self.load_collection(client, collection_name)
|
|
# 6. search
|
|
vectors_to_search = rng.random((1, default_dim))
|
|
insert_ids = [i for i in range(default_nb)]
|
|
self.search(client, collection_name, vectors_to_search,
|
|
check_task=CheckTasks.check_search_results,
|
|
check_items={"enable_milvus_client_api": True,
|
|
"nq": len(vectors_to_search),
|
|
"ids": insert_ids,
|
|
"limit": default_limit,
|
|
"pk_name": default_primary_key_field_name})
|
|
# 7. query
|
|
self.query(client, collection_name, filter=default_search_exp,
|
|
check_task=CheckTasks.check_query_results,
|
|
check_items={exp_res: rows,
|
|
"with_vec": True,
|
|
"pk_name": default_primary_key_field_name})
|
|
self.drop_collection(client, collection_name)
|
|
|
|
@pytest.mark.tags(CaseLabel.L2)
|
|
@pytest.mark.parametrize("index", ct.all_index_types[:8])
|
|
def test_milvus_client_index_after_insert(self, index):
|
|
"""
|
|
target: test index after insert
|
|
method: create connection, collection, insert, index and search
|
|
expected: index/search/query successfully
|
|
"""
|
|
metric_type = "COSINE"
|
|
client = self._client()
|
|
collection_name = cf.gen_unique_str(prefix)
|
|
# 1. create collection
|
|
self.create_collection(client, collection_name, default_dim, consistency_level="Strong")
|
|
self.release_collection(client, collection_name)
|
|
self.drop_index(client, collection_name, "vector")
|
|
# 2. insert
|
|
rng = np.random.default_rng(seed=19530)
|
|
rows = [{default_primary_key_field_name: i, default_vector_field_name: list(rng.random((1, default_dim))[0]),
|
|
default_float_field_name: i * 1.0, default_string_field_name: str(i)} for i in range(default_nb)]
|
|
self.insert(client, collection_name, rows)
|
|
# 3. prepare index params
|
|
index_params = self.prepare_index_params(client)[0]
|
|
params = cf.get_index_params_params(index)
|
|
index_params.add_index(field_name="vector", index_type=index, metric_type=metric_type, params=params)
|
|
# 4. create index
|
|
self.create_index(client, collection_name, index_params)
|
|
# 5. load collection
|
|
self.load_collection(client, collection_name)
|
|
# 5. search
|
|
vectors_to_search = rng.random((1, default_dim))
|
|
insert_ids = [i for i in range(default_nb)]
|
|
self.search(client, collection_name, vectors_to_search,
|
|
check_task=CheckTasks.check_search_results,
|
|
check_items={"enable_milvus_client_api": True,
|
|
"nq": len(vectors_to_search),
|
|
"ids": insert_ids,
|
|
"limit": default_limit,
|
|
"pk_name": default_primary_key_field_name})
|
|
# 4. query
|
|
self.query(client, collection_name, filter=default_search_exp,
|
|
check_task=CheckTasks.check_query_results,
|
|
check_items={exp_res: rows,
|
|
"with_vec": True,
|
|
"pk_name": default_primary_key_field_name})
|
|
self.drop_collection(client, collection_name)
|
|
|
|
@pytest.mark.tags(CaseLabel.L1)
|
|
@pytest.mark.parametrize("add_field", [True, False])
|
|
def test_milvus_client_scalar_auto_index(self, add_field):
|
|
"""
|
|
target: test index with autoindex on both scalar and vector field
|
|
method: create connection, collection, insert and search
|
|
expected: index/search/query successfully
|
|
"""
|
|
metric_type = "COSINE"
|
|
numeric_fields = [ct.default_int32_field_name, ct.default_int16_field_name,
|
|
ct.default_int8_field_name, default_float_field_name,
|
|
ct.default_double_field_name, ct.default_int64_field_name]
|
|
client = self._client()
|
|
collection_name = cf.gen_unique_str(prefix)
|
|
# 1. create collection
|
|
schema = self.create_schema(client)[0]
|
|
schema.add_field(default_primary_key_field_name, DataType.INT64, is_primary=True)
|
|
for field_name in numeric_fields:
|
|
schema.add_field(field_name, DataType.INT32, nullable=True)
|
|
for index in ct.varchar_supported_index_types:
|
|
schema.add_field(f"{default_string_field_name}_{index}", DataType.VARCHAR, max_length=64, nullable=True)
|
|
schema.add_field(ct.default_bool_field_name, DataType.BOOL)
|
|
schema.add_field(default_vector_field_name, DataType.FLOAT_VECTOR, dim=default_dim)
|
|
self.create_collection(client, collection_name, schema=schema, consistency_level="Strong")
|
|
if add_field:
|
|
self.add_collection_field(client, collection_name, field_name="field_int", data_type=DataType.INT32,
|
|
nullable=True)
|
|
self.add_collection_field(client, collection_name, field_name="field_varchar", data_type=DataType.VARCHAR,
|
|
nullable=True, max_length=64)
|
|
self.release_collection(client, collection_name)
|
|
self.drop_index(client, collection_name, "vector")
|
|
res = self.list_indexes(client, collection_name)[0]
|
|
assert res == []
|
|
# 2. prepare index params
|
|
index = "AUTOINDEX"
|
|
index_params = self.prepare_index_params(client)[0]
|
|
index_params.add_index(field_name=default_primary_key_field_name,
|
|
index_type=ct.numeric_supported_index_types[0], metric_type=metric_type)
|
|
index_params.add_index(field_name=default_vector_field_name, index_type=index, metric_type=metric_type)
|
|
index_params.add_index(field_name=ct.default_bool_field_name,
|
|
index_type="", metric_type=metric_type)
|
|
if len(numeric_fields) >= len(ct.numeric_supported_index_types):
|
|
k = 0
|
|
for i in range(len(numeric_fields)):
|
|
if k >= len(ct.numeric_supported_index_types):
|
|
k = 0
|
|
index_params.add_index(field_name=numeric_fields[i],
|
|
index_type=ct.numeric_supported_index_types[k], metric_type=metric_type)
|
|
k += 1
|
|
else:
|
|
k = 0
|
|
for i in range(len(ct.numeric_supported_index_types)):
|
|
if k >= len(numeric_fields):
|
|
k = 0
|
|
index_params.add_index(field_name=numeric_fields[k],
|
|
index_type=ct.numeric_supported_index_types[i], metric_type=metric_type)
|
|
k += 1
|
|
|
|
for index in ct.varchar_supported_index_types:
|
|
index_params.add_index(field_name=f"{default_string_field_name}_{index}",
|
|
index_type=index, metric_type=metric_type)
|
|
|
|
if add_field:
|
|
index_params.add_index(field_name="field_int",
|
|
index_type=ct.numeric_supported_index_types[0], metric_type=metric_type)
|
|
index_params.add_index(field_name="field_varchar",
|
|
index_type=ct.varchar_supported_index_types[0], metric_type=metric_type)
|
|
# 3. create index
|
|
self.create_index(client, collection_name, index_params)
|
|
# 4. drop index
|
|
self.drop_index(client, collection_name, default_vector_field_name)
|
|
for field_name in numeric_fields:
|
|
self.drop_index(client, collection_name, field_name)
|
|
for index in ct.varchar_supported_index_types:
|
|
self.drop_index(client, collection_name, f"{default_string_field_name}_{index}")
|
|
self.drop_index(client, collection_name, ct.default_bool_field_name)
|
|
self.drop_index(client, collection_name, default_primary_key_field_name)
|
|
if add_field:
|
|
self.drop_index(client, collection_name, "field_int")
|
|
self.drop_index(client, collection_name, "field_varchar")
|
|
# 5. create index
|
|
self.create_index(client, collection_name, index_params)
|
|
# 6. insert
|
|
collection_info = self.describe_collection(client, collection_name)[0]
|
|
rows = cf.gen_row_data_by_schema(nb=2000, schema=collection_info, start=0)
|
|
self.insert(client, collection_name, rows)
|
|
# 7. load collection
|
|
self.load_collection(client, collection_name)
|
|
# 8. search
|
|
vectors_to_search = cf.gen_vectors(nb=1, dim=default_dim)
|
|
insert_ids = [i for i in range(default_nb)]
|
|
filter_fields = []
|
|
filter_fields.extend(numeric_fields)
|
|
if add_field:
|
|
filter_fields.extend(["field_int", "field_varchar"])
|
|
for field_name in filter_fields:
|
|
self.search(client, collection_name, vectors_to_search,
|
|
limit=default_limit,
|
|
filter=f"{field_name} is null",
|
|
check_task=CheckTasks.check_search_results,
|
|
check_items={"enable_milvus_client_api": True,
|
|
"nq": len(vectors_to_search),
|
|
"ids": insert_ids,
|
|
"limit": default_limit,
|
|
"pk_name": default_primary_key_field_name})
|
|
for index in ct.varchar_supported_index_types:
|
|
self.search(client, collection_name, vectors_to_search,
|
|
limit=default_limit,
|
|
filter=f"{default_string_field_name}_{index} is not null",
|
|
check_task=CheckTasks.check_search_results,
|
|
check_items={"enable_milvus_client_api": True,
|
|
"nq": len(vectors_to_search),
|
|
"ids": insert_ids,
|
|
"limit": default_limit,
|
|
"pk_name": default_primary_key_field_name})
|
|
# 9. query
|
|
self.query(client, collection_name, filter=default_search_exp,
|
|
check_task=CheckTasks.check_query_results,
|
|
check_items={exp_res: rows,
|
|
"with_vec": True,
|
|
"pk_name": default_primary_key_field_name})
|
|
self.drop_collection(client, collection_name)
|
|
|
|
@pytest.mark.tags(CaseLabel.L1)
|
|
def test_milvus_client_scalar_hybrid_index_small_distinct_before_insert(self):
|
|
"""
|
|
target: test index with autoindex on int/varchar with small distinct value (<=100)
|
|
method: create connection, collection, insert and search
|
|
expected: index/search/query successfully (autoindex is bitmap index indeed)
|
|
"""
|
|
metric_type = "IP"
|
|
client = self._client()
|
|
collection_name = cf.gen_unique_str(prefix)
|
|
# 1. create collection
|
|
int64_field_name = "int"
|
|
schema = self.create_schema(client)[0]
|
|
schema.add_field(default_primary_key_field_name, DataType.INT64, is_primary=True)
|
|
schema.add_field(default_string_field_name, DataType.VARCHAR, max_length=64)
|
|
schema.add_field(int64_field_name, DataType.INT64)
|
|
schema.add_field(default_vector_field_name, DataType.FLOAT_VECTOR, dim=default_dim)
|
|
self.create_collection(client, collection_name, schema=schema, consistency_level="Strong")
|
|
self.release_collection(client, collection_name)
|
|
self.drop_index(client, collection_name, "vector")
|
|
res = self.list_indexes(client, collection_name)[0]
|
|
assert res == []
|
|
# 2. prepare index params
|
|
index = "AUTOINDEX"
|
|
index_params = self.prepare_index_params(client)[0]
|
|
index_params.add_index(field_name=default_vector_field_name, index_type=index, metric_type=metric_type)
|
|
index_params.add_index(field_name=int64_field_name, index_type=index, metric_type=metric_type)
|
|
index_params.add_index(field_name=default_string_field_name, index_type=index, metric_type=metric_type)
|
|
# 3. create index
|
|
self.create_index(client, collection_name, index_params)
|
|
# 4. insert
|
|
rng = np.random.default_rng(seed=19530)
|
|
rows = [{default_primary_key_field_name: i, default_vector_field_name: list(rng.random((1, default_dim))[0]),
|
|
int64_field_name: np.random.randint(0, 99), default_string_field_name: str(np.random.randint(0, 99))}
|
|
for i in range(default_nb)]
|
|
self.insert(client, collection_name, rows)
|
|
# 5. load collection
|
|
self.load_collection(client, collection_name)
|
|
# 6. search
|
|
vectors_to_search = rng.random((1, default_dim))
|
|
insert_ids = [i for i in range(default_nb)]
|
|
self.search(client, collection_name, vectors_to_search,
|
|
check_task=CheckTasks.check_search_results,
|
|
check_items={"enable_milvus_client_api": True,
|
|
"nq": len(vectors_to_search),
|
|
"ids": insert_ids,
|
|
"limit": default_limit,
|
|
"pk_name": default_primary_key_field_name})
|
|
# 7. query
|
|
self.query(client, collection_name, filter=default_search_exp,
|
|
check_task=CheckTasks.check_query_results,
|
|
check_items={exp_res: rows,
|
|
"with_vec": True,
|
|
"pk_name": default_primary_key_field_name})
|
|
self.drop_collection(client, collection_name)
|
|
|
|
@pytest.mark.tags(CaseLabel.L1)
|
|
def test_milvus_client_scalar_hybrid_index_small_to_large_distinct_after_insert(self):
|
|
"""
|
|
target: test index with autoindex on int/varchar with small distinct value (<=100) first and
|
|
insert to large distinct (2000+) later
|
|
method: create connection, collection, insert and search
|
|
expected: index/search/query successfully
|
|
"""
|
|
metric_type = "COSINE"
|
|
client = self._client()
|
|
collection_name = cf.gen_unique_str(prefix)
|
|
# 1. create collection
|
|
int64_field_name = "int"
|
|
schema = self.create_schema(client)[0]
|
|
schema.add_field(default_primary_key_field_name, DataType.INT64, is_primary=True)
|
|
schema.add_field(ct.default_int32_field_name, DataType.INT32)
|
|
schema.add_field(ct.default_int16_field_name, DataType.INT16)
|
|
schema.add_field(ct.default_int8_field_name, DataType.INT8)
|
|
schema.add_field(default_string_field_name, DataType.VARCHAR, max_length=64)
|
|
schema.add_field(int64_field_name, DataType.INT64)
|
|
schema.add_field(default_vector_field_name, DataType.FLOAT_VECTOR, dim=default_dim)
|
|
self.create_collection(client, collection_name, schema=schema, consistency_level="Strong")
|
|
self.release_collection(client, collection_name)
|
|
self.drop_index(client, collection_name, "vector")
|
|
res = self.list_indexes(client, collection_name)[0]
|
|
assert res == []
|
|
# 2. insert
|
|
rng = np.random.default_rng(seed=19530)
|
|
rows = [{default_primary_key_field_name: i, default_vector_field_name: list(rng.random((1, default_dim))[0]),
|
|
int64_field_name: np.random.randint(0, 99), ct.default_int32_field_name: np.int32(i),
|
|
ct.default_int16_field_name: np.int16(i), ct.default_int8_field_name: np.int8(i),
|
|
default_string_field_name: str(np.random.randint(0, 99))} for i in range(default_nb)]
|
|
self.insert(client, collection_name, rows)
|
|
# 3. prepare index params
|
|
index = "AUTOINDEX"
|
|
index_params = self.prepare_index_params(client)[0]
|
|
index_params.add_index(field_name=default_vector_field_name, index_type=index, metric_type=metric_type)
|
|
index_params.add_index(field_name=int64_field_name, index_type=index, metric_type=metric_type)
|
|
index_params.add_index(field_name=ct.default_int32_field_name, index_type="", metric_type=metric_type)
|
|
index_params.add_index(field_name=ct.default_int16_field_name, metric_type=metric_type)
|
|
index_params.add_index(field_name=ct.default_int8_field_name, index_type=index, metric_type=metric_type)
|
|
index_params.add_index(field_name=default_string_field_name, index_type=index, metric_type=metric_type)
|
|
# 4. create index
|
|
self.create_index(client, collection_name, index_params)
|
|
# 5. load collection
|
|
self.load_collection(client, collection_name)
|
|
# 6. search
|
|
vectors_to_search = rng.random((1, default_dim))
|
|
insert_ids = [i for i in range(default_nb)]
|
|
self.search(client, collection_name, vectors_to_search,
|
|
check_task=CheckTasks.check_search_results,
|
|
check_items={"enable_milvus_client_api": True,
|
|
"nq": len(vectors_to_search),
|
|
"ids": insert_ids,
|
|
"limit": default_limit,
|
|
"pk_name": default_primary_key_field_name})
|
|
# 7. query
|
|
self.query(client, collection_name, filter=default_search_exp,
|
|
check_task=CheckTasks.check_query_results,
|
|
check_items={exp_res: rows,
|
|
"with_vec": True,
|
|
"pk_name": default_primary_key_field_name})
|
|
# 8. insert more distinct value to the scalar field to make the autoindex change
|
|
rng = np.random.default_rng(seed=19530)
|
|
rows = [{default_primary_key_field_name: i, default_vector_field_name: list(rng.random((1, default_dim))[0]),
|
|
int64_field_name: np.random.randint(0, 99), ct.default_int32_field_name: np.int32(i),
|
|
ct.default_int16_field_name: np.int16(i), ct.default_int8_field_name: np.int8(i),
|
|
default_string_field_name: str(np.random.randint(0, 99))} for i in range(default_nb, 2 * default_nb)]
|
|
self.insert(client, collection_name, rows)
|
|
self.flush(client, collection_name)
|
|
# 9. search
|
|
vectors_to_search = rng.random((1, default_dim))
|
|
insert_ids = [i for i in range(2 * default_nb)]
|
|
self.search(client, collection_name, vectors_to_search,
|
|
check_task=CheckTasks.check_search_results,
|
|
check_items={"enable_milvus_client_api": True,
|
|
"nq": len(vectors_to_search),
|
|
"ids": insert_ids,
|
|
"limit": default_limit,
|
|
"pk_name": default_primary_key_field_name})
|
|
self.drop_collection(client, collection_name)
|
|
|
|
@pytest.mark.tags(CaseLabel.L2)
|
|
def test_milvus_client_vector_auto_index(self, metric_type):
|
|
"""
|
|
target: test index for multiple vectors
|
|
method: create connection, collection, index, insert and search
|
|
expected: index/search/query successfully
|
|
"""
|
|
client = self._client()
|
|
collection_name = cf.gen_unique_str(prefix)
|
|
# 1. create collection
|
|
self.create_collection(client, collection_name, default_dim, consistency_level="Strong")
|
|
self.release_collection(client, collection_name)
|
|
self.drop_index(client, collection_name, "vector")
|
|
res = self.list_indexes(client, collection_name)[0]
|
|
assert res == []
|
|
# 2. prepare index params
|
|
index = "AUTOINDEX"
|
|
numeric_index = "STL_SORT"
|
|
index_params = self.prepare_index_params(client)[0]
|
|
index_params.add_index(field_name="vector", index_type=index, metric_type=metric_type)
|
|
index_params.add_index(field_name="id", index_type=numeric_index, metric_type=metric_type)
|
|
# 3. create index
|
|
self.create_index(client, collection_name, index_params)
|
|
# 4. insert
|
|
rng = np.random.default_rng(seed=19530)
|
|
rows = [{default_primary_key_field_name: i, default_vector_field_name: list(rng.random((1, default_dim))[0]),
|
|
default_float_field_name: i * 1.0, default_string_field_name: str(i),
|
|
default_multiple_vector_field_name: list(rng.random((1, default_dim))[0])} for i in range(default_nb)]
|
|
self.insert(client, collection_name, rows)
|
|
# 5. load collection
|
|
self.load_collection(client, collection_name)
|
|
# 6. search
|
|
vectors_to_search = rng.random((1, default_dim))
|
|
insert_ids = [i for i in range(default_nb)]
|
|
self.search(client, collection_name, vectors_to_search,
|
|
check_task=CheckTasks.check_search_results,
|
|
check_items={"enable_milvus_client_api": True,
|
|
"nq": len(vectors_to_search),
|
|
"ids": insert_ids,
|
|
"limit": default_limit,
|
|
"pk_name": default_primary_key_field_name})
|
|
# 7. query
|
|
self.query(client, collection_name, filter=default_search_exp,
|
|
check_task=CheckTasks.check_query_results,
|
|
check_items={exp_res: rows,
|
|
"with_vec": True,
|
|
"pk_name": default_primary_key_field_name})
|
|
self.drop_collection(client, collection_name)
|
|
|
|
@pytest.mark.tags(CaseLabel.L2)
|
|
def test_milvus_client_index_drop_create_same_index(self):
|
|
"""
|
|
target: test index after drop and create same index twice
|
|
method: create connection, collection, create/drop/create index, insert and search
|
|
expected: index create/drop and search/query successfully
|
|
"""
|
|
client = self._client()
|
|
collection_name = cf.gen_unique_str(prefix)
|
|
# 1. create collection
|
|
self.create_collection(client, collection_name, default_dim, consistency_level="Strong")
|
|
self.release_collection(client, collection_name)
|
|
self.drop_index(client, collection_name, "vector")
|
|
res = self.list_indexes(client, collection_name)[0]
|
|
assert res == []
|
|
# 2. prepare index params
|
|
index_params = self.prepare_index_params(client)[0]
|
|
index_params.add_index(field_name="vector", index_type="HNSW", metric_type="L2")
|
|
# 3. create index
|
|
self.create_index(client, collection_name, index_params)
|
|
# 4. drop index
|
|
self.drop_index(client, collection_name, "vector")
|
|
# 4. create same index twice
|
|
self.create_index(client, collection_name, index_params)
|
|
# 5. insert
|
|
rng = np.random.default_rng(seed=19530)
|
|
rows = [{default_primary_key_field_name: i, default_vector_field_name: list(rng.random((1, default_dim))[0]),
|
|
default_float_field_name: i * 1.0, default_string_field_name: str(i)} for i in range(default_nb)]
|
|
self.insert(client, collection_name, rows)
|
|
# 6. load collection
|
|
self.load_collection(client, collection_name)
|
|
# 7. search
|
|
vectors_to_search = rng.random((1, default_dim))
|
|
insert_ids = [i for i in range(default_nb)]
|
|
self.search(client, collection_name, vectors_to_search,
|
|
check_task=CheckTasks.check_search_results,
|
|
check_items={"enable_milvus_client_api": True,
|
|
"nq": len(vectors_to_search),
|
|
"ids": insert_ids,
|
|
"limit": default_limit,
|
|
"pk_name": default_primary_key_field_name})
|
|
# 8. query
|
|
self.query(client, collection_name, filter=default_search_exp,
|
|
check_task=CheckTasks.check_query_results,
|
|
check_items={exp_res: rows,
|
|
"with_vec": True,
|
|
"pk_name": default_primary_key_field_name})
|
|
self.drop_collection(client, collection_name)
|
|
|
|
|
|
class TestMilvusClientJsonPathIndexInvalid(TestMilvusClientV2Base):
|
|
""" Test case of search interface """
|
|
|
|
@pytest.fixture(scope="function", params=["TRIE", "STL_SORT", "BITMAP"])
|
|
def not_supported_varchar_scalar_index(self, request):
|
|
yield request.param
|
|
|
|
@pytest.fixture(scope="function", params=["INVERTED"])
|
|
def supported_varchar_scalar_index(self, request):
|
|
yield request.param
|
|
|
|
@pytest.fixture(scope="function", params=[DataType.INT8.name, DataType.INT16.name, DataType.INT32.name,
|
|
DataType.INT64.name, DataType.FLOAT.name,
|
|
DataType.ARRAY.name, DataType.FLOAT_VECTOR.name,
|
|
DataType.FLOAT16_VECTOR.name, DataType.BFLOAT16_VECTOR.name,
|
|
DataType.BINARY_VECTOR.name,
|
|
DataType.SPARSE_FLOAT_VECTOR.name, DataType.INT8_VECTOR.name])
|
|
def not_supported_json_cast_type(self, request):
|
|
yield request.param
|
|
|
|
@pytest.fixture(scope="function", params=["Json", "BOOL", "double", "varchar"])
|
|
def supported_json_cast_type(self, request):
|
|
yield request.param
|
|
|
|
"""
|
|
******************************************************************
|
|
# The following are invalid base cases
|
|
******************************************************************
|
|
"""
|
|
|
|
@pytest.mark.tags(CaseLabel.L1)
|
|
def test_milvus_client_json_path_index_no_index_params(self):
|
|
"""
|
|
target: test json path index with:
|
|
1. no all index params
|
|
2. no json_cast_type
|
|
3. no json_path
|
|
method: create index on invalid collection name
|
|
expected: raise exception
|
|
"""
|
|
client = self._client()
|
|
collection_name = cf.gen_unique_str(prefix)
|
|
# 1. create collection
|
|
self.create_collection(client, collection_name, default_dim, consistency_level="Strong")
|
|
self.release_collection(client, collection_name)
|
|
self.drop_index(client, collection_name, "vector")
|
|
# 2. prepare index params with no index params
|
|
index_params = self.prepare_index_params(client)[0]
|
|
index_params.add_index(field_name=default_vector_field_name, index_type="AUTOINDEX", metric_type="COSINE")
|
|
index_params.add_index(field_name="my_json", index_type="INVERTED")
|
|
# 3. create index
|
|
error = {ct.err_code: 1100, ct.err_msg: "json index must specify cast type: missing parameter"
|
|
"[missing_param=json_cast_type]: invalid parameter"
|
|
"[expected=valid index params][actual=invalid index params]"}
|
|
self.create_index(client, collection_name, index_params,
|
|
check_task=CheckTasks.err_res, check_items=error)
|
|
# 4. prepare index params with no json_cast_type
|
|
index_params = self.prepare_index_params(client)[0]
|
|
index_params.add_index(field_name=default_vector_field_name, index_type="AUTOINDEX", metric_type="COSINE")
|
|
index_params.add_index(field_name="my_json", index_type="INVERTED", params={"json_path": "my_json['a']['b']"})
|
|
# 5. create index
|
|
error = {ct.err_code: 1100, ct.err_msg: "json index must specify cast type: missing parameter"
|
|
"[missing_param=json_cast_type]: invalid parameter"
|
|
"[expected=valid index params][actual=invalid index params]"}
|
|
self.create_index(client, collection_name, index_params,
|
|
check_task=CheckTasks.err_res, check_items=error)
|
|
# 6. prepare index params with no json_path
|
|
index_params = self.prepare_index_params(client)[0]
|
|
index_params.add_index(field_name=default_vector_field_name, index_type="AUTOINDEX", metric_type="COSINE")
|
|
index_params.add_index(field_name="my_json", index_type="INVERTED", params={"json_cast_type": "varchar"})
|
|
# 7. create index
|
|
self.create_index(client, collection_name, index_params)
|
|
self.drop_collection(client, collection_name)
|
|
|
|
@pytest.mark.tags(CaseLabel.L1)
|
|
@pytest.mark.parametrize("invalid_index_type", ["12-s", "12 s", "(mn)", "中文", "%$#"])
|
|
def test_milvus_client_json_path_index_invalid_index_type(self, invalid_index_type):
|
|
"""
|
|
target: test json path index with invalid index type
|
|
method: create json path index with invalid index type
|
|
expected: raise exception
|
|
"""
|
|
client = self._client()
|
|
collection_name = cf.gen_unique_str(prefix)
|
|
# 1. create collection
|
|
self.create_collection(client, collection_name, default_dim, consistency_level="Strong")
|
|
self.release_collection(client, collection_name)
|
|
self.drop_index(client, collection_name, "vector")
|
|
# 2. prepare index params with invalid index type
|
|
index_params = self.prepare_index_params(client)[0]
|
|
index_params.add_index(field_name=default_vector_field_name, index_type="AUTOINDEX", metric_type="COSINE")
|
|
index_params.add_index(field_name="my_json", index_type=invalid_index_type, params={"json_cast_type": "double",
|
|
"json_path": "my_json['a']['b']"})
|
|
# 3. create index
|
|
error = {ct.err_code: 1100, ct.err_msg: f"invalid parameter[expected=valid index]"
|
|
f"[actual=invalid index type: {invalid_index_type}]"}
|
|
self.create_index(client, collection_name, index_params,
|
|
check_task=CheckTasks.err_res, check_items=error)
|
|
|
|
@pytest.mark.tags(CaseLabel.L1)
|
|
@pytest.mark.parametrize("enable_dynamic_field", [True, False])
|
|
def test_milvus_client_json_path_index_not_support_index_type(self, enable_dynamic_field,
|
|
not_supported_varchar_scalar_index):
|
|
"""
|
|
target: test json path index with not supported index type
|
|
method: create json path index with not supported index type
|
|
expected: raise exception
|
|
"""
|
|
client = self._client()
|
|
collection_name = cf.gen_unique_str(prefix)
|
|
# 1. create collection
|
|
json_field_name = "my_json"
|
|
schema = self.create_schema(client, enable_dynamic_field=enable_dynamic_field)[0]
|
|
schema.add_field(default_primary_key_field_name, DataType.INT64, is_primary=True, auto_id=False)
|
|
schema.add_field(default_vector_field_name, DataType.FLOAT_VECTOR, dim=default_dim)
|
|
schema.add_field(default_string_field_name, DataType.VARCHAR, max_length=64)
|
|
if not enable_dynamic_field:
|
|
schema.add_field(json_field_name, DataType.JSON)
|
|
index_params = self.prepare_index_params(client)[0]
|
|
index_params.add_index(default_vector_field_name, metric_type="COSINE")
|
|
self.create_collection(client, collection_name, default_dim)
|
|
# 2. prepare index params with invalid json index type
|
|
index_params = self.prepare_index_params(client)[0]
|
|
index_params.add_index(field_name=default_vector_field_name, index_type="AUTOINDEX", metric_type="COSINE")
|
|
index_params.add_index(field_name=json_field_name, index_type=not_supported_varchar_scalar_index,
|
|
params={"json_cast_type": "DOUBLE", "json_path": "my_json['a']['b']"})
|
|
# 3. create index
|
|
if not_supported_varchar_scalar_index == "TRIE":
|
|
supported_field_type = "varchar"
|
|
got_json_suffix = ""
|
|
if not_supported_varchar_scalar_index == "STL_SORT":
|
|
supported_field_type = "numeric, varchar or timestamptz"
|
|
got_json_suffix = ", got JSON"
|
|
if not_supported_varchar_scalar_index == "BITMAP":
|
|
supported_field_type = "bool, int, string and array"
|
|
not_supported_varchar_scalar_index = "bitmap index"
|
|
got_json_suffix = ""
|
|
error = {ct.err_code: 1100, ct.err_msg: f"{not_supported_varchar_scalar_index} are only supported on "
|
|
f"{supported_field_type} field{got_json_suffix}: invalid parameter[expected=valid "
|
|
f"index params][actual=invalid index params]"}
|
|
self.create_index(client, collection_name, index_params,
|
|
check_task=CheckTasks.err_res, check_items=error)
|
|
|
|
@pytest.mark.tags(CaseLabel.L1)
|
|
@pytest.mark.parametrize("enable_dynamic_field", [True, False])
|
|
@pytest.mark.parametrize("invalid_json_cast_type", ["12-s", "12 s", "(mn)", "中文", "%$#", 1, 1.0])
|
|
def test_milvus_client_json_path_index_invalid_json_cast_type(self, enable_dynamic_field, invalid_json_cast_type,
|
|
supported_varchar_scalar_index):
|
|
"""
|
|
target: test json path index with invalid json_cast_type
|
|
method: create json path index with invalid json_cast_type
|
|
expected: raise exception
|
|
"""
|
|
client = self._client()
|
|
collection_name = cf.gen_unique_str(prefix)
|
|
# 1. create collection
|
|
json_field_name = "my_json"
|
|
schema = self.create_schema(client, enable_dynamic_field=enable_dynamic_field)[0]
|
|
schema.add_field(default_primary_key_field_name, DataType.INT64, is_primary=True, auto_id=False)
|
|
schema.add_field(default_vector_field_name, DataType.FLOAT_VECTOR, dim=default_dim)
|
|
schema.add_field(default_string_field_name, DataType.VARCHAR, max_length=64)
|
|
if not enable_dynamic_field:
|
|
schema.add_field(json_field_name, DataType.JSON)
|
|
index_params = self.prepare_index_params(client)[0]
|
|
index_params.add_index(default_vector_field_name, metric_type="COSINE")
|
|
self.create_collection(client, collection_name, default_dim)
|
|
# 2. prepare index params with invalid json index type
|
|
index_params = self.prepare_index_params(client)[0]
|
|
index_params.add_index(field_name=default_vector_field_name, index_type="AUTOINDEX", metric_type="COSINE")
|
|
index_params.add_index(field_name=json_field_name, index_name="json_index",
|
|
index_type=supported_varchar_scalar_index,
|
|
params={"json_cast_type": invalid_json_cast_type,
|
|
"json_path": f"{json_field_name}['a']['b']"})
|
|
# 3. create index
|
|
error = {ct.err_code: 1100, ct.err_msg: f"index params][actual=invalid index params]"}
|
|
self.create_index(client, collection_name, index_params,
|
|
check_task=CheckTasks.err_res, check_items=error)
|
|
|
|
@pytest.mark.tags(CaseLabel.L1)
|
|
@pytest.mark.parametrize("enable_dynamic_field", [True])
|
|
def test_milvus_client_json_path_index_not_supported_json_cast_type(self, enable_dynamic_field,
|
|
supported_varchar_scalar_index):
|
|
"""
|
|
target: test json path index with not supported json_cast_type
|
|
method: create json path index with not supported json_cast_type
|
|
expected: raise exception
|
|
"""
|
|
client = self._client()
|
|
collection_name = cf.gen_unique_str(prefix)
|
|
# 1. create collection
|
|
json_field_name = "my_json"
|
|
schema = self.create_schema(client, enable_dynamic_field=enable_dynamic_field)[0]
|
|
schema.add_field(default_primary_key_field_name, DataType.INT64, is_primary=True, auto_id=False)
|
|
schema.add_field(default_vector_field_name, DataType.FLOAT_VECTOR, dim=default_dim)
|
|
schema.add_field(default_string_field_name, DataType.VARCHAR, max_length=64)
|
|
if not enable_dynamic_field:
|
|
schema.add_field(json_field_name, DataType.JSON)
|
|
index_params = self.prepare_index_params(client)[0]
|
|
index_params.add_index(default_vector_field_name, metric_type="COSINE")
|
|
self.create_collection(client, collection_name, default_dim)
|
|
# 2. prepare index params with invalid json index type
|
|
for cast_type in ct.not_supported_json_cast_types:
|
|
index_params = self.prepare_index_params(client)[0]
|
|
index_params.add_index(field_name=default_vector_field_name, index_type="AUTOINDEX", metric_type="COSINE")
|
|
index_params.add_index(field_name=json_field_name, index_name="json_index",
|
|
index_type=supported_varchar_scalar_index,
|
|
params={"json_cast_type": cast_type,
|
|
"json_path": f"{json_field_name}['a']['b']"})
|
|
# 3. create index
|
|
error = {ct.err_code: 1100, ct.err_msg: f"index params][actual=invalid index params]"}
|
|
self.create_index(client, collection_name, index_params,
|
|
check_task=CheckTasks.err_res, check_items=error)
|
|
|
|
@pytest.mark.tags(CaseLabel.L1)
|
|
@pytest.mark.parametrize("enable_dynamic_field", [False])
|
|
@pytest.mark.parametrize("invalid_json_path", [1, 1.0, '/'])
|
|
def test_milvus_client_json_path_index_invalid_json_path(self, enable_dynamic_field, invalid_json_path,
|
|
supported_varchar_scalar_index):
|
|
"""
|
|
target: test json path index with invalid json_cast_type
|
|
method: create json path index with invalid json_cast_type
|
|
expected: raise exception
|
|
"""
|
|
client = self._client()
|
|
collection_name = cf.gen_unique_str(prefix)
|
|
# 1. create collection
|
|
json_field_name = "my_json"
|
|
schema = self.create_schema(client, enable_dynamic_field=enable_dynamic_field)[0]
|
|
schema.add_field(default_primary_key_field_name, DataType.INT64, is_primary=True, auto_id=False)
|
|
schema.add_field(default_vector_field_name, DataType.FLOAT_VECTOR, dim=default_dim)
|
|
schema.add_field(default_string_field_name, DataType.VARCHAR, max_length=64)
|
|
if not enable_dynamic_field:
|
|
schema.add_field(json_field_name, DataType.JSON)
|
|
index_params = self.prepare_index_params(client)[0]
|
|
index_params.add_index(default_vector_field_name, metric_type="COSINE")
|
|
self.create_collection(client, collection_name, default_dim)
|
|
# 2. prepare index params with invalid json index type
|
|
index_params = self.prepare_index_params(client)[0]
|
|
index_params.add_index(field_name=default_vector_field_name, index_type="AUTOINDEX", metric_type="COSINE")
|
|
index_params.add_index(field_name=json_field_name, index_name="json_index",
|
|
index_type=supported_varchar_scalar_index,
|
|
params={"json_cast_type": "Double", "json_path": invalid_json_path})
|
|
# 3. create index
|
|
error = {ct.err_code: 65535, ct.err_msg: f"cannot parse identifier: {invalid_json_path}"}
|
|
self.create_index(client, collection_name, index_params,
|
|
check_task=CheckTasks.err_res, check_items=error)
|
|
|
|
@pytest.mark.tags(CaseLabel.L1)
|
|
def test_milvus_client_json_path_index_not_exist_field_non_dynamic(self, supported_varchar_scalar_index):
|
|
"""
|
|
target: test json path index with not exist field in non dynamic field scenario
|
|
method: create json path index with not exist field with enable_dynamic_field disabled
|
|
expected: raise exception
|
|
"""
|
|
client = self._client()
|
|
collection_name = cf.gen_unique_str(prefix)
|
|
# 1. create collection
|
|
json_field_name = "my_json"
|
|
schema = self.create_schema(client, enable_dynamic_field=False)[0]
|
|
schema.add_field(default_primary_key_field_name, DataType.INT64, is_primary=True, auto_id=False)
|
|
schema.add_field(default_vector_field_name, DataType.FLOAT_VECTOR, dim=default_dim)
|
|
schema.add_field(default_string_field_name, DataType.VARCHAR, max_length=64)
|
|
index_params = self.prepare_index_params(client)[0]
|
|
index_params.add_index(field_name=default_vector_field_name, index_type="AUTOINDEX", metric_type="COSINE")
|
|
index_params.add_index(field_name=json_field_name, index_type=supported_varchar_scalar_index,
|
|
params={"json_cast_type": "double", "json_path": f"{json_field_name}['a']"})
|
|
error = {ct.err_code: 65535, ct.err_msg: f"cannot create index on non-exist field: {json_field_name}"}
|
|
self.create_collection(client, collection_name, schema=schema, index_params=index_params,
|
|
check_task=CheckTasks.err_res, check_items=error)
|
|
|
|
@pytest.mark.tags(CaseLabel.L1)
|
|
@pytest.mark.parametrize("enable_dynamic_field", [True, False])
|
|
def test_milvus_client_different_index_same_json_path(self, enable_dynamic_field, supported_varchar_scalar_index):
|
|
"""
|
|
target: test create different index with different json_cast_type on the same json path of the same field
|
|
method: create different index with different json_cast_type on the same
|
|
json path of the same field (same index name)
|
|
expected: raise exception
|
|
"""
|
|
client = self._client()
|
|
collection_name = cf.gen_unique_str(prefix)
|
|
# 1. create collection
|
|
json_field_name = "my_json"
|
|
schema = self.create_schema(client, enable_dynamic_field=enable_dynamic_field)[0]
|
|
schema.add_field(default_primary_key_field_name, DataType.INT64, is_primary=True, auto_id=False)
|
|
schema.add_field(default_vector_field_name, DataType.FLOAT_VECTOR, dim=default_dim)
|
|
schema.add_field(default_string_field_name, DataType.VARCHAR, max_length=64)
|
|
if not enable_dynamic_field:
|
|
schema.add_field(json_field_name, DataType.JSON)
|
|
index_params = self.prepare_index_params(client)[0]
|
|
index_params.add_index(default_vector_field_name, metric_type="COSINE")
|
|
self.create_collection(client, collection_name, default_dim)
|
|
# 2. prepare index params
|
|
index_params = self.prepare_index_params(client)[0]
|
|
index_params.add_index(field_name=default_vector_field_name, index_type="AUTOINDEX", metric_type="COSINE")
|
|
index_params.add_index(field_name=json_field_name, index_name="json_index",
|
|
index_type=supported_varchar_scalar_index,
|
|
params={"json_cast_type": "double", "json_path": f"{json_field_name}['a']"})
|
|
self.create_index(client, collection_name, index_params)
|
|
# 4. prepare another index params
|
|
index_params = self.prepare_index_params(client)[0]
|
|
index_params.add_index(field_name=json_field_name, index_name="json_index",
|
|
index_type=supported_varchar_scalar_index,
|
|
params={"json_cast_type": "varchar", "json_path": f"{json_field_name}['a']"})
|
|
# 5. create index
|
|
error = {ct.err_code: 65535, ct.err_msg: "CreateIndex failed: at most one distinct index is allowed per field"}
|
|
self.create_index(client, collection_name, index_params,
|
|
check_task=CheckTasks.err_res, check_items=error)
|
|
|
|
@pytest.mark.tags(CaseLabel.L1)
|
|
@pytest.mark.parametrize("enable_dynamic_field", [True, False])
|
|
def test_milvus_client_different_index_name_same_json_path(self, enable_dynamic_field,
|
|
supported_varchar_scalar_index):
|
|
"""
|
|
target: test json path index with different index name but with same json path
|
|
method: create json path index with different index name but with same json path
|
|
expected: raise exception
|
|
"""
|
|
client = self._client()
|
|
collection_name = cf.gen_unique_str(prefix)
|
|
# 1. create collection
|
|
json_field_name = "my_json"
|
|
schema = self.create_schema(client, enable_dynamic_field=enable_dynamic_field)[0]
|
|
schema.add_field(default_primary_key_field_name, DataType.INT64, is_primary=True, auto_id=False)
|
|
schema.add_field(default_vector_field_name, DataType.FLOAT_VECTOR, dim=default_dim)
|
|
schema.add_field(default_string_field_name, DataType.VARCHAR, max_length=64)
|
|
if not enable_dynamic_field:
|
|
schema.add_field(json_field_name, DataType.JSON)
|
|
index_params = self.prepare_index_params(client)[0]
|
|
index_params.add_index(default_vector_field_name, metric_type="COSINE")
|
|
self.create_collection(client, collection_name, default_dim)
|
|
# 2. prepare index params
|
|
index_params = self.prepare_index_params(client)[0]
|
|
index_params.add_index(field_name=default_vector_field_name, index_type="AUTOINDEX", metric_type="COSINE")
|
|
index_params.add_index(field_name=json_field_name, index_name="json_index_1",
|
|
index_type=supported_varchar_scalar_index,
|
|
params={"json_cast_type": "varchar", "json_path": f"{json_field_name}['a']"})
|
|
self.create_index(client, collection_name, index_params)
|
|
# 4. prepare another index params
|
|
index_params = self.prepare_index_params(client)[0]
|
|
index_params.add_index(field_name=json_field_name, index_name="json_index_2",
|
|
index_type=supported_varchar_scalar_index,
|
|
params={"json_cast_type": "varchar", "json_path": f"{json_field_name}['a']"})
|
|
# 5. create index
|
|
error = {ct.err_code: 65535, ct.err_msg: "CreateIndex failed: creating multiple "
|
|
"indexes on same field is not supported"}
|
|
self.create_index(client, collection_name, index_params,
|
|
check_task=CheckTasks.err_res, check_items=error)
|
|
|
|
@pytest.mark.tags(CaseLabel.L1)
|
|
@pytest.mark.parametrize("enable_dynamic_field", [True, False])
|
|
def test_milvus_client_different_json_path_index_same_field_same_index_name(self, enable_dynamic_field,
|
|
supported_json_cast_type,
|
|
supported_varchar_scalar_index):
|
|
"""
|
|
target: test different json path index with same index name at the same time
|
|
method: test different json path index with same index name at the same index_params object
|
|
expected: raise exception
|
|
"""
|
|
client = self._client()
|
|
collection_name = cf.gen_unique_str(prefix)
|
|
# 1. create collection
|
|
json_field_name = "my_json"
|
|
schema = self.create_schema(client, enable_dynamic_field=enable_dynamic_field)[0]
|
|
schema.add_field(default_primary_key_field_name, DataType.INT64, is_primary=True, auto_id=False)
|
|
schema.add_field(default_vector_field_name, DataType.FLOAT_VECTOR, dim=default_dim)
|
|
schema.add_field(default_string_field_name, DataType.VARCHAR, max_length=64)
|
|
if not enable_dynamic_field:
|
|
schema.add_field(json_field_name, DataType.JSON)
|
|
index_params = self.prepare_index_params(client)[0]
|
|
index_params.add_index(default_vector_field_name, metric_type="COSINE")
|
|
self.create_collection(client, collection_name, schema=schema, index_params=index_params)
|
|
# 2. insert
|
|
vectors = cf.gen_vectors(default_nb, default_dim)
|
|
rows = [{default_primary_key_field_name: i, default_vector_field_name: vectors[i],
|
|
default_string_field_name: str(i), json_field_name: {'a': {"b": i}}} for i in range(default_nb)]
|
|
self.insert(client, collection_name, rows)
|
|
# 3. prepare index params
|
|
index_name = "json_index"
|
|
index_params = self.prepare_index_params(client)[0]
|
|
index_params.add_index(field_name=default_vector_field_name, index_type="AUTOINDEX", metric_type="COSINE")
|
|
index_params.add_index(field_name=json_field_name, index_name=index_name,
|
|
index_type=supported_varchar_scalar_index,
|
|
params={"json_cast_type": supported_json_cast_type,
|
|
"json_path": f"{json_field_name}['a']['b']"})
|
|
index_params.add_index(field_name=json_field_name, index_name=index_name,
|
|
index_type=supported_varchar_scalar_index,
|
|
params={"json_cast_type": supported_json_cast_type,
|
|
"json_path": f"{json_field_name}['a']"})
|
|
index_params.add_index(field_name=json_field_name, index_name=index_name,
|
|
index_type=supported_varchar_scalar_index,
|
|
params={"json_cast_type": supported_json_cast_type,
|
|
"json_path": f"{json_field_name}"})
|
|
# 4. create index
|
|
error = {ct.err_code: 65535, ct.err_msg: "CreateIndex failed: at most one distinct index is allowed per field"}
|
|
self.create_index(client, collection_name, index_params,
|
|
check_task=CheckTasks.err_res, check_items=error)
|
|
|
|
|
|
class TestMilvusClientJsonPathIndexValid(TestMilvusClientV2Base):
|
|
""" Test case of search interface """
|
|
|
|
# @pytest.fixture(scope="function", params=["TRIE", "STL_SORT", "BITMAP"])
|
|
# def not_supported_varchar_scalar_index(self, request):
|
|
# yield request.param
|
|
|
|
@pytest.fixture(scope="function", params=["INVERTED"])
|
|
def supported_varchar_scalar_index(self, request):
|
|
yield request.param
|
|
|
|
@pytest.fixture(scope="function", params=["DOUBLE", "VARCHAR", "BOOL", "double", "varchar", "bool"])
|
|
def supported_json_cast_type(self, request):
|
|
yield request.param
|
|
|
|
"""
|
|
******************************************************************
|
|
# The following are valid base cases
|
|
******************************************************************
|
|
"""
|
|
|
|
@pytest.mark.tags(CaseLabel.L0)
|
|
@pytest.mark.parametrize("enable_dynamic_field", [True, False])
|
|
def test_milvus_client_json_path_index_default(self, enable_dynamic_field, supported_json_cast_type,
|
|
supported_varchar_scalar_index):
|
|
"""
|
|
target: test json path index with default parameter
|
|
method: create json path index with default parameter
|
|
expected: create json path index successfully
|
|
"""
|
|
client = self._client()
|
|
collection_name = cf.gen_unique_str(prefix)
|
|
# 1. create collection
|
|
json_field_name = "my_json"
|
|
schema = self.create_schema(client, enable_dynamic_field=enable_dynamic_field)[0]
|
|
schema.add_field(default_primary_key_field_name, DataType.INT64, is_primary=True, auto_id=False)
|
|
schema.add_field(default_vector_field_name, DataType.FLOAT_VECTOR, dim=default_dim)
|
|
schema.add_field(default_string_field_name, DataType.VARCHAR, max_length=64)
|
|
if not enable_dynamic_field:
|
|
schema.add_field(json_field_name, DataType.JSON)
|
|
index_params = self.prepare_index_params(client)[0]
|
|
index_params.add_index(default_vector_field_name, metric_type="COSINE")
|
|
self.create_collection(client, collection_name, schema=schema, index_params=index_params)
|
|
# 2. insert with different data distribution
|
|
vectors = cf.gen_vectors(default_nb + 50, default_dim)
|
|
rows = [{default_primary_key_field_name: i, default_vector_field_name: vectors[i],
|
|
default_string_field_name: str(i), json_field_name: {'a': {"b": i}}} for i in
|
|
range(default_nb)]
|
|
self.insert(client, collection_name, rows)
|
|
rows = [{default_primary_key_field_name: i, default_vector_field_name: vectors[i],
|
|
default_string_field_name: str(i), json_field_name: i} for i in
|
|
range(default_nb, default_nb + 10)]
|
|
self.insert(client, collection_name, rows)
|
|
rows = [{default_primary_key_field_name: i, default_vector_field_name: vectors[i],
|
|
default_string_field_name: str(i), json_field_name: {}} for i in
|
|
range(default_nb + 10, default_nb + 20)]
|
|
self.insert(client, collection_name, rows)
|
|
rows = [{default_primary_key_field_name: i, default_vector_field_name: vectors[i],
|
|
default_string_field_name: str(i), json_field_name: {'a': [1, 2, 3]}} for i in
|
|
range(default_nb + 20, default_nb + 30)]
|
|
self.insert(client, collection_name, rows)
|
|
rows = [{default_primary_key_field_name: i, default_vector_field_name: vectors[i],
|
|
default_string_field_name: str(i), json_field_name: {'a': [{'b': 1}, 2, 3]}} for i in
|
|
range(default_nb + 20, default_nb + 30)]
|
|
self.insert(client, collection_name, rows)
|
|
rows = [{default_primary_key_field_name: i, default_vector_field_name: vectors[i],
|
|
default_string_field_name: str(i), json_field_name: {'a': [{'b': None}, 2, 3]}} for i in
|
|
range(default_nb + 30, default_nb + 40)]
|
|
self.insert(client, collection_name, rows)
|
|
# 2. prepare index params
|
|
index_params = self.prepare_index_params(client)[0]
|
|
index_params.add_index(field_name=default_vector_field_name, index_type="AUTOINDEX", metric_type="COSINE")
|
|
index_params.add_index(field_name=json_field_name, index_type=supported_varchar_scalar_index,
|
|
params={"json_cast_type": supported_json_cast_type,
|
|
"json_path": f"{json_field_name}['a']['b']"})
|
|
index_params.add_index(field_name=json_field_name,
|
|
index_type=supported_varchar_scalar_index,
|
|
params={"json_cast_type": supported_json_cast_type,
|
|
"json_path": f"{json_field_name}['a']"})
|
|
index_params.add_index(field_name=json_field_name,
|
|
index_type=supported_varchar_scalar_index,
|
|
params={"json_cast_type": supported_json_cast_type,
|
|
"json_path": f"{json_field_name}"})
|
|
index_params.add_index(field_name=json_field_name,
|
|
index_type=supported_varchar_scalar_index,
|
|
params={"json_cast_type": supported_json_cast_type,
|
|
"json_path": f"{json_field_name}['a'][0]['b']"})
|
|
index_params.add_index(field_name=json_field_name,
|
|
index_type=supported_varchar_scalar_index,
|
|
params={"json_cast_type": supported_json_cast_type,
|
|
"json_path": f"{json_field_name}['a'][0]"})
|
|
# 3. create index
|
|
self.create_index(client, collection_name, index_params)
|
|
self.list_indexes(client, collection_name)
|
|
index_name = json_field_name
|
|
if enable_dynamic_field:
|
|
index_name = "$meta/" + json_field_name
|
|
self.describe_index(client, collection_name, index_name + "/a/b",
|
|
check_task=CheckTasks.check_describe_index_property,
|
|
check_items={
|
|
"json_cast_type": supported_json_cast_type,
|
|
"json_path": f"{json_field_name}['a']['b']",
|
|
"index_type": supported_varchar_scalar_index,
|
|
"field_name": json_field_name,
|
|
"index_name": index_name + "/a/b"})
|
|
# 5. create same json index twice
|
|
self.create_index(client, collection_name, index_params)
|
|
self.describe_index(client, collection_name, index_name + "/a/b",
|
|
check_task=CheckTasks.check_describe_index_property,
|
|
check_items={
|
|
"json_cast_type": supported_json_cast_type,
|
|
"json_path": f"{json_field_name}['a']['b']",
|
|
"index_type": supported_varchar_scalar_index,
|
|
"field_name": json_field_name,
|
|
"index_name": index_name + "/a/b"})
|
|
self.describe_index(client, collection_name, index_name + '/a',
|
|
check_task=CheckTasks.check_describe_index_property,
|
|
check_items={
|
|
"json_cast_type": supported_json_cast_type,
|
|
"json_path": f"{json_field_name}['a']",
|
|
"index_type": supported_varchar_scalar_index,
|
|
"field_name": json_field_name,
|
|
"index_name": index_name + '/a'})
|
|
self.describe_index(client, collection_name, index_name,
|
|
check_task=CheckTasks.check_describe_index_property,
|
|
check_items={
|
|
"json_cast_type": supported_json_cast_type,
|
|
"json_path": f"{json_field_name}",
|
|
"index_type": supported_varchar_scalar_index,
|
|
"field_name": json_field_name,
|
|
"index_name": index_name})
|
|
self.describe_index(client, collection_name, index_name + '/a/0/b',
|
|
check_task=CheckTasks.check_describe_index_property,
|
|
check_items={
|
|
"json_cast_type": supported_json_cast_type,
|
|
"json_path": f"{json_field_name}['a'][0]['b']",
|
|
"index_type": supported_varchar_scalar_index,
|
|
"field_name": json_field_name,
|
|
"index_name": index_name + '/a/0/b'})
|
|
self.describe_index(client, collection_name, index_name + '/a/0',
|
|
check_task=CheckTasks.check_describe_index_property,
|
|
check_items={
|
|
"json_cast_type": supported_json_cast_type,
|
|
"json_path": f"{json_field_name}['a'][0]",
|
|
"index_type": supported_varchar_scalar_index,
|
|
"field_name": json_field_name,
|
|
"index_name": index_name + '/a/0'})
|
|
|
|
@pytest.mark.tags(CaseLabel.L1)
|
|
@pytest.mark.parametrize("enable_dynamic_field", [True, False])
|
|
def test_milvus_client_json_path_index_default_index_name(self, enable_dynamic_field, supported_json_cast_type,
|
|
supported_varchar_scalar_index):
|
|
"""
|
|
target: test json path index with not supported json_cast_type
|
|
method: create json path index with not supported json_cast_type
|
|
expected: raise exception
|
|
"""
|
|
client = self._client()
|
|
collection_name = cf.gen_unique_str(prefix)
|
|
# 1. create collection
|
|
json_field_name = "my_json"
|
|
schema = self.create_schema(client, enable_dynamic_field=enable_dynamic_field)[0]
|
|
schema.add_field(default_primary_key_field_name, DataType.INT64, is_primary=True, auto_id=False)
|
|
schema.add_field(default_vector_field_name, DataType.FLOAT_VECTOR, dim=default_dim)
|
|
schema.add_field(default_string_field_name, DataType.VARCHAR, max_length=64)
|
|
if not enable_dynamic_field:
|
|
schema.add_field(json_field_name, DataType.JSON)
|
|
index_params = self.prepare_index_params(client)[0]
|
|
index_params.add_index(default_vector_field_name, metric_type="COSINE")
|
|
self.create_collection(client, collection_name, schema=schema, index_params=index_params)
|
|
# 2. insert
|
|
vectors = cf.gen_vectors(default_nb, default_dim)
|
|
rows = [{default_primary_key_field_name: i, default_vector_field_name: vectors[i],
|
|
default_string_field_name: str(i), json_field_name: {'a': {"b": i}}} for i in range(default_nb)]
|
|
self.insert(client, collection_name, rows)
|
|
# 2. prepare index params
|
|
index_params = self.prepare_index_params(client)[0]
|
|
index_params.add_index(field_name=default_vector_field_name, index_type="AUTOINDEX", metric_type="COSINE")
|
|
index_params.add_index(field_name=json_field_name, index_type=supported_varchar_scalar_index,
|
|
params={"json_cast_type": supported_json_cast_type,
|
|
"json_path": f"{json_field_name}['a']['b']"})
|
|
# 3. create index
|
|
if enable_dynamic_field:
|
|
index_name = "$meta/" + json_field_name + '/a/b'
|
|
else:
|
|
index_name = json_field_name + '/a/b'
|
|
self.create_index(client, collection_name, index_params)
|
|
self.describe_index(client, collection_name, index_name,
|
|
check_task=CheckTasks.check_describe_index_property,
|
|
check_items={
|
|
"json_cast_type": supported_json_cast_type,
|
|
"json_path": f"{json_field_name}['a']['b']",
|
|
"index_type": supported_varchar_scalar_index,
|
|
"field_name": json_field_name,
|
|
"index_name": index_name})
|
|
|
|
@pytest.mark.tags(CaseLabel.L2)
|
|
@pytest.mark.skip(reason="issue #40636")
|
|
def test_milvus_client_json_path_index_on_non_json_field(self, supported_json_cast_type,
|
|
supported_varchar_scalar_index):
|
|
"""
|
|
target: test json path index with "json_cast_type" and "json_path" parameters on non json field
|
|
method: create json path index with "json_cast_type" and "json_path" parameters on int64 field
|
|
steps: 1. create schema with id, vector and varchar fields
|
|
2. prepare index parameters with default vector index
|
|
3. create collection with the above defined schema and index params
|
|
4. insert default_nb numbers of data
|
|
5. prepare index params with "json_cast_type" and "json_path" params on int64 field
|
|
6. create index with the new index params
|
|
7. check that the results of describe_index interface does not contain the "json_cast_type" and "json_path" parameters
|
|
expected: create the original inverted index successfully
|
|
"""
|
|
client = self._client()
|
|
collection_name = cf.gen_unique_str(prefix)
|
|
# 1. create collection
|
|
schema = self.create_schema(client, enable_dynamic_field=False)[0]
|
|
schema.add_field(default_primary_key_field_name, DataType.INT64, is_primary=True, auto_id=False)
|
|
schema.add_field(default_vector_field_name, DataType.FLOAT_VECTOR, dim=default_dim)
|
|
schema.add_field(default_string_field_name, DataType.VARCHAR, max_length=64)
|
|
index_params = self.prepare_index_params(client)[0]
|
|
index_params.add_index(default_vector_field_name, metric_type="COSINE")
|
|
self.create_collection(client, collection_name, schema=schema, index_params=index_params)
|
|
# 2. insert
|
|
vectors = cf.gen_vectors(default_nb, default_dim)
|
|
rows = [{default_primary_key_field_name: i, default_vector_field_name: vectors[i],
|
|
default_string_field_name: str(i)} for i in range(default_nb)]
|
|
self.insert(client, collection_name, rows)
|
|
# 3. prepare index params
|
|
index_params = self.prepare_index_params(client)[0]
|
|
index_params.add_index(field_name=default_vector_field_name, index_type="AUTOINDEX", metric_type="COSINE")
|
|
index_params.add_index(field_name=default_string_field_name, index_type=supported_varchar_scalar_index,
|
|
params={"json_cast_type": supported_json_cast_type,
|
|
"json_path": f"{default_string_field_name}['a']['b']"})
|
|
# 4. create index
|
|
index_name = default_string_field_name
|
|
self.create_index(client, collection_name, index_params)
|
|
self.describe_index(client, collection_name, index_name,
|
|
check_task=CheckTasks.check_describe_index_property,
|
|
check_items={
|
|
"json_cast_type": supported_json_cast_type,
|
|
"json_path": f"{default_string_field_name}['a']['b']",
|
|
"index_type": supported_varchar_scalar_index,
|
|
"field_name": default_string_field_name,
|
|
"index_name": index_name})
|
|
|
|
@pytest.mark.tags(CaseLabel.L2)
|
|
@pytest.mark.parametrize("enable_dynamic_field", [True, False])
|
|
def test_milvus_client_different_json_path_index_same_field_different_index_name(self, enable_dynamic_field,
|
|
supported_json_cast_type,
|
|
supported_varchar_scalar_index):
|
|
"""
|
|
target: test different json path index with different default index name at the same time
|
|
method: test different json path index with different default index name at the same index_params object
|
|
expected: create index successfully using the last index params with the same index name
|
|
"""
|
|
client = self._client()
|
|
collection_name = cf.gen_unique_str(prefix)
|
|
# 1. create collection
|
|
json_field_name = "my_json"
|
|
schema = self.create_schema(client, enable_dynamic_field=enable_dynamic_field)[0]
|
|
schema.add_field(default_primary_key_field_name, DataType.INT64, is_primary=True, auto_id=False)
|
|
schema.add_field(default_vector_field_name, DataType.FLOAT_VECTOR, dim=default_dim)
|
|
schema.add_field(default_string_field_name, DataType.VARCHAR, max_length=64)
|
|
if not enable_dynamic_field:
|
|
schema.add_field(json_field_name, DataType.JSON)
|
|
index_params = self.prepare_index_params(client)[0]
|
|
index_params.add_index(default_vector_field_name, metric_type="COSINE")
|
|
self.create_collection(client, collection_name, schema=schema, index_params=index_params)
|
|
# 2. insert
|
|
vectors = cf.gen_vectors(default_nb, default_dim)
|
|
rows = [{default_primary_key_field_name: i, default_vector_field_name: vectors[i],
|
|
default_string_field_name: str(i), json_field_name: {'a': {"b": i}}} for i in range(default_nb)]
|
|
self.insert(client, collection_name, rows)
|
|
# 2. prepare index params
|
|
index_name = "json_index"
|
|
index_params = self.prepare_index_params(client)[0]
|
|
index_params.add_index(field_name=default_vector_field_name, index_type="AUTOINDEX", metric_type="COSINE")
|
|
index_params.add_index(field_name=json_field_name, index_name=index_name + "1",
|
|
index_type=supported_varchar_scalar_index,
|
|
params={"json_cast_type": supported_json_cast_type,
|
|
"json_path": f"{json_field_name}['a']['b']"})
|
|
index_params.add_index(field_name=json_field_name, index_name=index_name + "2",
|
|
index_type=supported_varchar_scalar_index,
|
|
params={"json_cast_type": supported_json_cast_type,
|
|
"json_path": f"{json_field_name}['a']"})
|
|
index_params.add_index(field_name=json_field_name, index_name=index_name + "3",
|
|
index_type=supported_varchar_scalar_index,
|
|
params={"json_cast_type": supported_json_cast_type,
|
|
"json_path": f"{json_field_name}"})
|
|
# 3. create index
|
|
self.create_index(client, collection_name, index_params)
|
|
self.describe_index(client, collection_name, index_name + '1',
|
|
check_task=CheckTasks.check_describe_index_property,
|
|
check_items={
|
|
"json_cast_type": supported_json_cast_type,
|
|
"json_path": f"{json_field_name}['a']['b']",
|
|
"index_type": supported_varchar_scalar_index,
|
|
"field_name": json_field_name,
|
|
"index_name": index_name + '1'})
|
|
self.describe_index(client, collection_name, index_name + '2',
|
|
check_task=CheckTasks.check_describe_index_property,
|
|
check_items={
|
|
"json_cast_type": supported_json_cast_type,
|
|
"json_path": f"{json_field_name}['a']",
|
|
"index_type": supported_varchar_scalar_index,
|
|
"field_name": json_field_name,
|
|
"index_name": index_name + '2'})
|
|
self.describe_index(client, collection_name, index_name + '3',
|
|
check_task=CheckTasks.check_describe_index_property,
|
|
check_items={
|
|
"json_cast_type": supported_json_cast_type,
|
|
"json_path": f"{json_field_name}",
|
|
"index_type": supported_varchar_scalar_index,
|
|
"field_name": json_field_name,
|
|
"index_name": index_name + '3'})
|
|
|
|
@pytest.mark.tags(CaseLabel.L1)
|
|
@pytest.mark.parametrize("enable_dynamic_field", [True, False])
|
|
def test_milvus_client_diff_index_same_field_diff_index_name_diff_index_params(self, enable_dynamic_field,
|
|
supported_json_cast_type,
|
|
supported_varchar_scalar_index):
|
|
"""
|
|
target: test different json path index with different default index name at the same time
|
|
method: test different json path index with different default index name at different index_params object
|
|
expected: create index successfully with all the indexes created
|
|
"""
|
|
client = self._client()
|
|
collection_name = cf.gen_unique_str(prefix)
|
|
# 1. create collection
|
|
json_field_name = "my_json"
|
|
schema = self.create_schema(client, enable_dynamic_field=enable_dynamic_field)[0]
|
|
schema.add_field(default_primary_key_field_name, DataType.INT64, is_primary=True, auto_id=False)
|
|
schema.add_field(default_vector_field_name, DataType.FLOAT_VECTOR, dim=default_dim)
|
|
schema.add_field(default_string_field_name, DataType.VARCHAR, max_length=64)
|
|
if not enable_dynamic_field:
|
|
schema.add_field(json_field_name, DataType.JSON)
|
|
index_params = self.prepare_index_params(client)[0]
|
|
index_params.add_index(default_vector_field_name, metric_type="COSINE")
|
|
self.create_collection(client, collection_name, schema=schema, index_params=index_params)
|
|
self.load_collection(client, collection_name)
|
|
# 2. insert
|
|
vectors = cf.gen_vectors(default_nb, default_dim)
|
|
rows = [{default_primary_key_field_name: i, default_vector_field_name: vectors[i],
|
|
default_string_field_name: str(i), json_field_name: {'a': {"b": i}}} for i in range(default_nb)]
|
|
self.insert(client, collection_name, rows)
|
|
# 2. prepare index params
|
|
index_params = self.prepare_index_params(client)[0]
|
|
index_params.add_index(field_name=json_field_name, index_type=supported_varchar_scalar_index,
|
|
params={"json_cast_type": supported_json_cast_type,
|
|
"json_path": f"{json_field_name}['a']['b']"})
|
|
self.create_index(client, collection_name, index_params)
|
|
index_params = self.prepare_index_params(client)[0]
|
|
index_params.add_index(field_name=json_field_name,
|
|
index_type=supported_varchar_scalar_index,
|
|
params={"json_cast_type": supported_json_cast_type,
|
|
"json_path": f"{json_field_name}['a']"})
|
|
self.create_index(client, collection_name, index_params)
|
|
index_params = self.prepare_index_params(client)[0]
|
|
index_params.add_index(field_name=json_field_name,
|
|
index_type=supported_varchar_scalar_index,
|
|
params={"json_cast_type": supported_json_cast_type,
|
|
"json_path": f"{json_field_name}"})
|
|
self.create_index(client, collection_name, index_params)
|
|
# 3. create index
|
|
index_name = f"{json_field_name}/a/b"
|
|
if enable_dynamic_field:
|
|
index_name = "$meta/" + index_name
|
|
self.describe_index(client, collection_name, index_name,
|
|
check_task=CheckTasks.check_describe_index_property,
|
|
check_items={
|
|
"json_cast_type": supported_json_cast_type,
|
|
"json_path": "my_json['a']['b']",
|
|
"index_type": supported_varchar_scalar_index,
|
|
"field_name": json_field_name,
|
|
"index_name": index_name})
|
|
index_name = f"{json_field_name}/a"
|
|
if enable_dynamic_field:
|
|
index_name = "$meta/" + index_name
|
|
self.describe_index(client, collection_name, index_name,
|
|
check_task=CheckTasks.check_describe_index_property,
|
|
check_items={
|
|
"json_cast_type": supported_json_cast_type,
|
|
"json_path": "my_json['a']",
|
|
"index_type": supported_varchar_scalar_index,
|
|
"field_name": json_field_name,
|
|
"index_name": index_name})
|
|
index_name = f"{json_field_name}"
|
|
if enable_dynamic_field:
|
|
index_name = "$meta/" + index_name
|
|
self.describe_index(client, collection_name, index_name,
|
|
check_task=CheckTasks.check_describe_index_property,
|
|
check_items={
|
|
"json_cast_type": supported_json_cast_type,
|
|
"json_path": "my_json",
|
|
"index_type": supported_varchar_scalar_index,
|
|
"field_name": json_field_name,
|
|
"index_name": index_name
|
|
})
|
|
|
|
@pytest.mark.tags(CaseLabel.L2)
|
|
@pytest.mark.parametrize("enable_dynamic_field", [True, False])
|
|
def test_milvus_client_json_index_same_json_path_diff_field(self, enable_dynamic_field, supported_json_cast_type,
|
|
supported_varchar_scalar_index):
|
|
"""
|
|
target: test different json path index with different default index name at the same time
|
|
method: test different json path index with different default index name at the same index_params object
|
|
expected: create index successfully using the last index params with the same index name
|
|
"""
|
|
client = self._client()
|
|
collection_name = cf.gen_unique_str(prefix)
|
|
# 1. create collection
|
|
json_field_name = "my_json"
|
|
schema = self.create_schema(client, enable_dynamic_field=enable_dynamic_field)[0]
|
|
schema.add_field(default_primary_key_field_name, DataType.INT64, is_primary=True, auto_id=False)
|
|
schema.add_field(default_vector_field_name, DataType.FLOAT_VECTOR, dim=default_dim)
|
|
schema.add_field(default_string_field_name, DataType.VARCHAR, max_length=64)
|
|
if not enable_dynamic_field:
|
|
schema.add_field(json_field_name, DataType.JSON)
|
|
schema.add_field(json_field_name + "1", DataType.JSON)
|
|
index_params = self.prepare_index_params(client)[0]
|
|
index_params.add_index(default_vector_field_name, metric_type="COSINE")
|
|
self.create_collection(client, collection_name, schema=schema, index_params=index_params)
|
|
# 2. insert
|
|
vectors = cf.gen_vectors(default_nb, default_dim)
|
|
rows = [{default_primary_key_field_name: i, default_vector_field_name: vectors[i],
|
|
default_string_field_name: str(i), json_field_name: {'a': {'b': i}},
|
|
json_field_name + "1": {'a': {'b': i}}} for i in range(default_nb)]
|
|
self.insert(client, collection_name, rows)
|
|
# 2. prepare index params
|
|
index_params = self.prepare_index_params(client)[0]
|
|
index_params.add_index(field_name=json_field_name, index_type=supported_varchar_scalar_index,
|
|
params={"json_cast_type": supported_json_cast_type,
|
|
"json_path": f"{json_field_name}['a']['b']"})
|
|
self.create_index(client, collection_name, index_params)
|
|
index_params = self.prepare_index_params(client)[0]
|
|
index_params.add_index(field_name=json_field_name + "1",
|
|
index_type=supported_varchar_scalar_index,
|
|
params={"json_cast_type": supported_json_cast_type,
|
|
"json_path": f"{json_field_name}1['a']['b']"})
|
|
self.create_index(client, collection_name, index_params)
|
|
# 3. create index
|
|
index_name = f"{json_field_name}/a/b"
|
|
if enable_dynamic_field:
|
|
index_name = "$meta/" + index_name
|
|
self.describe_index(client, collection_name, index_name,
|
|
check_task=CheckTasks.check_describe_index_property,
|
|
check_items={
|
|
"json_cast_type": supported_json_cast_type,
|
|
"json_path": f"{json_field_name}['a']['b']",
|
|
"index_type": supported_varchar_scalar_index,
|
|
"field_name": json_field_name,
|
|
"index_name": index_name})
|
|
index_name = f"{json_field_name}1/a/b"
|
|
if enable_dynamic_field:
|
|
index_name = "$meta/" + index_name
|
|
self.describe_index(client, collection_name, index_name,
|
|
check_task=CheckTasks.check_describe_index_property,
|
|
check_items={
|
|
"json_cast_type": supported_json_cast_type,
|
|
"json_path": f"{json_field_name}1['a']['b']",
|
|
"index_type": supported_varchar_scalar_index,
|
|
"field_name": json_field_name + "1",
|
|
"index_name": index_name})
|
|
|
|
@pytest.mark.tags(CaseLabel.L2)
|
|
@pytest.mark.parametrize("enable_dynamic_field", [True, False])
|
|
def test_milvus_client_json_path_index_before_load(self, enable_dynamic_field, supported_json_cast_type,
|
|
supported_varchar_scalar_index):
|
|
"""
|
|
target: test json path index with not supported json_cast_type
|
|
method: create json path index with not supported json_cast_type
|
|
expected: raise exception
|
|
"""
|
|
client = self._client()
|
|
collection_name = cf.gen_unique_str(prefix)
|
|
# 1. create collection
|
|
json_field_name = "my_json"
|
|
schema = self.create_schema(client, enable_dynamic_field=enable_dynamic_field)[0]
|
|
schema.add_field(default_primary_key_field_name, DataType.INT64, is_primary=True, auto_id=False)
|
|
schema.add_field(default_vector_field_name, DataType.FLOAT_VECTOR, dim=default_dim)
|
|
schema.add_field(default_string_field_name, DataType.VARCHAR, max_length=64)
|
|
if not enable_dynamic_field:
|
|
schema.add_field(json_field_name, DataType.JSON)
|
|
index_params = self.prepare_index_params(client)[0]
|
|
index_params.add_index(default_vector_field_name, metric_type="COSINE")
|
|
self.create_collection(client, collection_name, schema=schema, index_params=index_params)
|
|
# 2. release collection
|
|
self.release_collection(client, collection_name)
|
|
# 3. insert with different data distribution
|
|
vectors = cf.gen_vectors(default_nb + 50, default_dim)
|
|
rows = [{default_primary_key_field_name: i, default_vector_field_name: vectors[i],
|
|
default_string_field_name: str(i), json_field_name: {'a': {"b": i}}} for i in
|
|
range(default_nb)]
|
|
self.insert(client, collection_name, rows)
|
|
rows = [{default_primary_key_field_name: i, default_vector_field_name: vectors[i],
|
|
default_string_field_name: str(i), json_field_name: i} for i in
|
|
range(default_nb, default_nb + 10)]
|
|
self.insert(client, collection_name, rows)
|
|
rows = [{default_primary_key_field_name: i, default_vector_field_name: vectors[i],
|
|
default_string_field_name: str(i), json_field_name: {}} for i in
|
|
range(default_nb + 10, default_nb + 20)]
|
|
self.insert(client, collection_name, rows)
|
|
rows = [{default_primary_key_field_name: i, default_vector_field_name: vectors[i],
|
|
default_string_field_name: str(i), json_field_name: {'a': [1, 2, 3]}} for i in
|
|
range(default_nb + 20, default_nb + 30)]
|
|
self.insert(client, collection_name, rows)
|
|
rows = [{default_primary_key_field_name: i, default_vector_field_name: vectors[i],
|
|
default_string_field_name: str(i), json_field_name: {'a': [{'b': 1}, 2, 3]}} for i in
|
|
range(default_nb + 20, default_nb + 30)]
|
|
self.insert(client, collection_name, rows)
|
|
rows = [{default_primary_key_field_name: i, default_vector_field_name: vectors[i],
|
|
default_string_field_name: str(i), json_field_name: {'a': [{'b': None}, 2, 3]}} for i in
|
|
range(default_nb + 30, default_nb + 40)]
|
|
self.insert(client, collection_name, rows)
|
|
# 4. prepare index params
|
|
index_name = "json_index"
|
|
index_params = self.prepare_index_params(client)[0]
|
|
index_params.add_index(field_name=default_vector_field_name, index_type="AUTOINDEX", metric_type="COSINE")
|
|
index_params.add_index(field_name=json_field_name, index_name=index_name,
|
|
index_type=supported_varchar_scalar_index,
|
|
params={"json_cast_type": supported_json_cast_type,
|
|
"json_path": f"{json_field_name}['a']['b']"})
|
|
index_params.add_index(field_name=json_field_name, index_name=index_name + '1',
|
|
index_type=supported_varchar_scalar_index,
|
|
params={"json_cast_type": supported_json_cast_type,
|
|
"json_path": f"{json_field_name}['a']"})
|
|
index_params.add_index(field_name=json_field_name, index_name=index_name + '2',
|
|
index_type=supported_varchar_scalar_index,
|
|
params={"json_cast_type": supported_json_cast_type,
|
|
"json_path": f"{json_field_name}"})
|
|
index_params.add_index(field_name=json_field_name, index_name=index_name + '3',
|
|
index_type=supported_varchar_scalar_index,
|
|
params={"json_cast_type": supported_json_cast_type,
|
|
"json_path": f"{json_field_name}['a'][0]['b']"})
|
|
index_params.add_index(field_name=json_field_name, index_name=index_name + '4',
|
|
index_type=supported_varchar_scalar_index,
|
|
params={"json_cast_type": supported_json_cast_type,
|
|
"json_path": f"{json_field_name}['a'][0]"})
|
|
# 5. create index
|
|
self.create_index(client, collection_name, index_params)
|
|
self.describe_index(client, collection_name, index_name,
|
|
check_task=CheckTasks.check_describe_index_property,
|
|
check_items={
|
|
# "json_cast_type": supported_json_cast_type, # issue 40426
|
|
"json_path": f"{json_field_name}['a']['b']",
|
|
"index_type": supported_varchar_scalar_index,
|
|
"field_name": json_field_name,
|
|
"index_name": index_name})
|
|
# 6. create json index on different json path
|
|
self.create_index(client, collection_name, index_params)
|
|
# 7. create same json index twice
|
|
self.create_index(client, collection_name, index_params)
|
|
self.describe_index(client, collection_name, index_name,
|
|
check_task=CheckTasks.check_describe_index_property,
|
|
check_items={
|
|
# "json_cast_type": supported_json_cast_type, # issue 40426
|
|
"json_path": f"{json_field_name}['a']['b']",
|
|
"index_type": supported_varchar_scalar_index,
|
|
"field_name": json_field_name,
|
|
"index_name": index_name})
|
|
self.describe_index(client, collection_name, index_name + '1',
|
|
check_task=CheckTasks.check_describe_index_property,
|
|
check_items={
|
|
# "json_cast_type": supported_json_cast_type, # issue 40426
|
|
"json_path": f"{json_field_name}['a']",
|
|
"index_type": supported_varchar_scalar_index,
|
|
"field_name": json_field_name,
|
|
"index_name": index_name + '1'})
|
|
self.describe_index(client, collection_name, index_name + '2',
|
|
check_task=CheckTasks.check_describe_index_property,
|
|
check_items={
|
|
# "json_cast_type": supported_json_cast_type, # issue 40426
|
|
"json_path": f"{json_field_name}",
|
|
"index_type": supported_varchar_scalar_index,
|
|
"field_name": json_field_name,
|
|
"index_name": index_name + '2'})
|
|
self.describe_index(client, collection_name, index_name + '3',
|
|
check_task=CheckTasks.check_describe_index_property,
|
|
check_items={
|
|
# "json_cast_type": supported_json_cast_type, # issue 40426
|
|
"json_path": f"{json_field_name}['a'][0]['b']",
|
|
"index_type": supported_varchar_scalar_index,
|
|
"field_name": json_field_name,
|
|
"index_name": index_name + '3'})
|
|
self.describe_index(client, collection_name, index_name + '4',
|
|
check_task=CheckTasks.check_describe_index_property,
|
|
check_items={
|
|
# "json_cast_type": supported_json_cast_type, # issue 40426
|
|
"json_path": f"{json_field_name}['a'][0]",
|
|
"index_type": supported_varchar_scalar_index,
|
|
"field_name": json_field_name,
|
|
"index_name": index_name + '4'})
|
|
|
|
@pytest.mark.tags(CaseLabel.L2)
|
|
def test_milvus_client_json_path_index_after_enable_dynamic_field(self, supported_json_cast_type,
|
|
supported_varchar_scalar_index):
|
|
"""
|
|
target: test json path index after enabling dynamic field
|
|
method: create collection, enable dynamic field, create json path index, insert
|
|
expected: json path index works with dynamic field data
|
|
"""
|
|
client = self._client()
|
|
collection_name = cf.gen_unique_str(prefix)
|
|
# 1. create collection
|
|
json_field_name = "dynamic_field"
|
|
schema = self.create_schema(client, enable_dynamic_field=False)[0]
|
|
schema.add_field(default_primary_key_field_name, DataType.INT64, is_primary=True, auto_id=False)
|
|
schema.add_field(default_vector_field_name, DataType.FLOAT_VECTOR, dim=default_dim)
|
|
schema.add_field(default_string_field_name, DataType.VARCHAR, max_length=64)
|
|
index_params = self.prepare_index_params(client)[0]
|
|
index_params.add_index(default_vector_field_name, metric_type="COSINE")
|
|
self.create_collection(client, collection_name, schema=schema, index_params=index_params)
|
|
# 2. alter collection dynamic field enable
|
|
self.alter_collection_properties(client, collection_name, {"dynamicfield.enabled": True})
|
|
res = self.describe_collection(client, collection_name)[0]
|
|
assert res.get('enable_dynamic_field', None) is True
|
|
# 3. insert with different data distribution
|
|
vectors = cf.gen_vectors(default_nb + 50, default_dim)
|
|
rows = [{default_primary_key_field_name: i, default_vector_field_name: vectors[i],
|
|
default_string_field_name: str(i), json_field_name: {'a': {"b": i}}} for i in range(default_nb)]
|
|
self.insert(client, collection_name, rows)
|
|
rows = [{default_primary_key_field_name: i, default_vector_field_name: vectors[i],
|
|
default_string_field_name: str(i), json_field_name: i} for i in range(default_nb, default_nb + 10)]
|
|
self.insert(client, collection_name, rows)
|
|
rows = [{default_primary_key_field_name: i, default_vector_field_name: vectors[i],
|
|
default_string_field_name: str(i), json_field_name: {}} for i in
|
|
range(default_nb + 10, default_nb + 20)]
|
|
self.insert(client, collection_name, rows)
|
|
rows = [{default_primary_key_field_name: i, default_vector_field_name: vectors[i],
|
|
default_string_field_name: str(i), json_field_name: {'a': [1, 2, 3]}}
|
|
for i in range(default_nb + 20, default_nb + 30)]
|
|
self.insert(client, collection_name, rows)
|
|
rows = [{default_primary_key_field_name: i, default_vector_field_name: vectors[i],
|
|
default_string_field_name: str(i), json_field_name: {'a': [{'b': 1}, 2, 3]}}
|
|
for i in range(default_nb + 20, default_nb + 30)]
|
|
self.insert(client, collection_name, rows)
|
|
rows = [{default_primary_key_field_name: i, default_vector_field_name: vectors[i],
|
|
default_string_field_name: str(i), json_field_name: {'a': [{'b': None}, 2, 3]}}
|
|
for i in range(default_nb + 30, default_nb + 40)]
|
|
self.insert(client, collection_name, rows)
|
|
# 4. prepare index params
|
|
index_params = self.prepare_index_params(client)[0]
|
|
index_params.add_index(field_name=default_vector_field_name, index_type="AUTOINDEX", metric_type="COSINE")
|
|
index_params.add_index(field_name=json_field_name, index_type=supported_varchar_scalar_index,
|
|
params={"json_cast_type": supported_json_cast_type,
|
|
"json_path": f"{json_field_name}['a']['b']"})
|
|
index_params.add_index(field_name=json_field_name,
|
|
index_type=supported_varchar_scalar_index,
|
|
params={"json_cast_type": supported_json_cast_type,
|
|
"json_path": f"{json_field_name}['a']"})
|
|
index_params.add_index(field_name=json_field_name,
|
|
index_type=supported_varchar_scalar_index,
|
|
params={"json_cast_type": supported_json_cast_type,
|
|
"json_path": f"{json_field_name}"})
|
|
index_params.add_index(field_name=json_field_name,
|
|
index_type=supported_varchar_scalar_index,
|
|
params={"json_cast_type": supported_json_cast_type,
|
|
"json_path": f"{json_field_name}['a'][0]['b']"})
|
|
index_params.add_index(field_name=json_field_name,
|
|
index_type=supported_varchar_scalar_index,
|
|
params={"json_cast_type": supported_json_cast_type,
|
|
"json_path": f"{json_field_name}['a'][0]"})
|
|
# 5. create index
|
|
self.create_index(client, collection_name, index_params)
|
|
self.list_indexes(client, collection_name)
|
|
index_name = "$meta/" + json_field_name
|
|
self.describe_index(client, collection_name, index_name + "/a/b",
|
|
check_task=CheckTasks.check_describe_index_property,
|
|
check_items={
|
|
"json_cast_type": supported_json_cast_type,
|
|
"json_path": f"{json_field_name}['a']['b']",
|
|
"index_type": supported_varchar_scalar_index,
|
|
"field_name": json_field_name,
|
|
"index_name": index_name + "/a/b"})
|
|
# 6. create same json index twice
|
|
self.create_index(client, collection_name, index_params)
|
|
self.describe_index(client, collection_name, index_name + "/a/b",
|
|
check_task=CheckTasks.check_describe_index_property,
|
|
check_items={
|
|
"json_cast_type": supported_json_cast_type,
|
|
"json_path": f"{json_field_name}['a']['b']",
|
|
"index_type": supported_varchar_scalar_index,
|
|
"field_name": json_field_name,
|
|
"index_name": index_name + "/a/b"})
|
|
self.describe_index(client, collection_name, index_name + '/a',
|
|
check_task=CheckTasks.check_describe_index_property,
|
|
check_items={
|
|
"json_cast_type": supported_json_cast_type,
|
|
"json_path": f"{json_field_name}['a']",
|
|
"index_type": supported_varchar_scalar_index,
|
|
"field_name": json_field_name,
|
|
"index_name": index_name + '/a'})
|
|
self.describe_index(client, collection_name, index_name,
|
|
check_task=CheckTasks.check_describe_index_property,
|
|
check_items={
|
|
"json_cast_type": supported_json_cast_type,
|
|
"json_path": f"{json_field_name}",
|
|
"index_type": supported_varchar_scalar_index,
|
|
"field_name": json_field_name,
|
|
"index_name": index_name})
|
|
self.describe_index(client, collection_name, index_name + '/a/0/b',
|
|
check_task=CheckTasks.check_describe_index_property,
|
|
check_items={
|
|
"json_cast_type": supported_json_cast_type,
|
|
"json_path": f"{json_field_name}['a'][0]['b']",
|
|
"index_type": supported_varchar_scalar_index,
|
|
"field_name": json_field_name,
|
|
"index_name": index_name + '/a/0/b'})
|
|
self.describe_index(client, collection_name, index_name + '/a/0',
|
|
check_task=CheckTasks.check_describe_index_property,
|
|
check_items={
|
|
"json_cast_type": supported_json_cast_type,
|
|
"json_path": f"{json_field_name}['a'][0]",
|
|
"index_type": supported_varchar_scalar_index,
|
|
"field_name": json_field_name,
|
|
"index_name": index_name + '/a/0'})
|