admin/milvus
1
0
Fork 0
mirror of https://gitee.com/milvus-io/milvus.git synced 2026-01-07 19:31:51 +08:00
Code Issues Packages Projects Releases Wiki Activity

test: add group by for fts hybrid search (#43037)

Browse Source 
/kind improvement

Signed-off-by: zhuwenxing <wenxing.zhu@zilliz.com>
This commit is contained in:
zhuwenxing 2025-07-02 11:16:50 +08:00 committed by GitHub
parent e8011908ac
commit 7b26cef3be
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
2 changed files with 17 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
tests/python_client/pytest.ini
View File

@ -6,10 +6,10 @@ addopts = --host localhost --html=/tmp/ci_logs/report.html --self-contained-html
log_format = [%(asctime)s - %(levelname)s - %(name)s]: %(message)s (%(filename)s:%(lineno)s)
log_date_format = %Y-%m-%d %H:%M:%S
markers =
tags: custom tags for test cases
filterwarnings =
ignore::DeprecationWarning
asyncio_default_fixture_loop_scope = function

17
tests/python_client/testcases/test_full_text_search.py
View File

@ -3536,6 +3536,7 @@ class TestHybridSearchWithFullTextSearch(TestcaseBase):
@pytest.mark.parametrize("empty_percent", [0])
@pytest.mark.parametrize("enable_partition_key", [True])
@pytest.mark.parametrize("enable_inverted_index", [True])
@pytest.mark.parametrize("enable_group_by_field", [True, False])
@pytest.mark.parametrize("index_type", ["SPARSE_INVERTED_INDEX"])
@pytest.mark.parametrize("tokenizer", ["standard"])
@pytest.mark.parametrize("inverted_index_algo", ct.inverted_index_algo)
@ -3547,6 +3548,7 @@ class TestHybridSearchWithFullTextSearch(TestcaseBase):
empty_percent,
index_type,
inverted_index_algo,
enable_group_by_field,
):
"""
target: test full text search
@ -3561,6 +3563,11 @@ class TestHybridSearchWithFullTextSearch(TestcaseBase):
dim = 128
fields = [
FieldSchema(name="id", dtype=DataType.INT64, is_primary=True),
FieldSchema(
name="language",
dtype=DataType.VARCHAR,
max_length=16,
),
FieldSchema(
name="word",
dtype=DataType.VARCHAR,
@ -3608,10 +3615,12 @@ class TestHybridSearchWithFullTextSearch(TestcaseBase):
collection_w = self.init_collection_wrap(
name=cf.gen_unique_str(prefix), schema=schema
)
language_list = ["en", "zh", "de", "jp", "unknown"]
fake = fake_en
data = [
{
"id": i,
"language": random.choice(language_list),
"word": fake.word().lower() if random.random() >= empty_percent else "",
"sentence": fake.sentence().lower()
if random.random() >= empty_percent
@ -3690,13 +3699,17 @@ class TestHybridSearchWithFullTextSearch(TestcaseBase):
reqs=[bm25_search, dense_search, sparse_search],
rerank=WeightedRanker(0.5, 0.5, 0.5),
limit=limit,
output_fields=["id", "text"],
output_fields=["id", "text", "language"],
group_by_field="language" if enable_group_by_field else None,
)
assert len(res_list) == nq
# check the result correctness
for i in range(nq):
log.info(f"res length: {len(res_list[i])}")
assert len(res_list[i]) == limit
if enable_group_by_field:
assert len(res_list[i]) == len(language_list)
else:
assert len(res_list[i]) == limit
class TestFullTextSearchMultiAnalyzer(TestcaseBase):