mirror of
https://github.com/qdrant/qdrant-client.git
synced 2026-07-28 13:41:11 -05:00
3f56da7d02
* update openapi and grpc + conversions * add strict mode to conversions + coverage * add strict mode to collection creation * add local mode for has_vectors condition * regen async with python 3.10 * tests for has-vector * fix filters in tests * fix conversion * fix test * Has vector tests (#878) * new: add has vector fixture * tests: add has vector multivector test * fix: add missing strict mode config usage and conversion (#877) * fix: add missing strict mode config usage and conversion * fix: pass strict mode config in update collection * fix: add strict mode config to recreate collection --------- Co-authored-by: George <george.panchuk@qdrant.tech>
86 lines
2.5 KiB
Python
86 lines
2.5 KiB
Python
from qdrant_client import models
|
|
|
|
from tests.congruence_tests.test_common import (
|
|
COLLECTION_NAME,
|
|
compare_client_results,
|
|
generate_fixtures,
|
|
generate_sparse_fixtures,
|
|
init_local,
|
|
init_client,
|
|
init_remote,
|
|
sparse_vectors_config,
|
|
generate_multivector_fixtures,
|
|
multi_vector_config,
|
|
)
|
|
|
|
|
|
def test_has_vector(local_client, remote_client):
|
|
points = generate_fixtures(100, skip_vectors=True)
|
|
|
|
local_client.upload_points(COLLECTION_NAME, points)
|
|
remote_client.upload_points(COLLECTION_NAME, points, wait=True)
|
|
|
|
local_client.upload_points(COLLECTION_NAME, points)
|
|
remote_client.upload_points(COLLECTION_NAME, points, wait=True)
|
|
|
|
compare_client_results(
|
|
local_client,
|
|
remote_client,
|
|
lambda c: c.scroll(
|
|
COLLECTION_NAME,
|
|
limit=50,
|
|
scroll_filter=models.Filter(must=[models.HasVectorCondition(has_vector="image")]),
|
|
)[0],
|
|
)
|
|
|
|
|
|
def test_has_vector_sparse():
|
|
points = generate_sparse_fixtures(100, skip_vectors=True)
|
|
|
|
local_client = init_local()
|
|
init_client(local_client, [], sparse_vectors_config=sparse_vectors_config)
|
|
|
|
remote_client = init_remote()
|
|
init_client(remote_client, [], sparse_vectors_config=sparse_vectors_config)
|
|
|
|
local_client.upload_points(COLLECTION_NAME, points)
|
|
remote_client.upload_points(COLLECTION_NAME, points, wait=True)
|
|
|
|
compare_client_results(
|
|
local_client,
|
|
remote_client,
|
|
lambda c: c.scroll(
|
|
COLLECTION_NAME,
|
|
limit=50,
|
|
scroll_filter=models.Filter(
|
|
must=[models.HasVectorCondition(has_vector="sparse-image")]
|
|
),
|
|
)[0],
|
|
)
|
|
|
|
|
|
def test_has_vector_multi():
|
|
points = generate_multivector_fixtures(100, skip_vectors=True)
|
|
|
|
local_client = init_local()
|
|
init_client(local_client, [], vectors_config=multi_vector_config)
|
|
|
|
remote_client = init_remote()
|
|
init_client(remote_client, [], vectors_config=multi_vector_config)
|
|
|
|
local_client.upload_points(COLLECTION_NAME, points)
|
|
remote_client.upload_points(COLLECTION_NAME, points, wait=True)
|
|
|
|
local_client.upload_points(COLLECTION_NAME, points)
|
|
remote_client.upload_points(COLLECTION_NAME, points, wait=True)
|
|
|
|
compare_client_results(
|
|
local_client,
|
|
remote_client,
|
|
lambda c: c.scroll(
|
|
COLLECTION_NAME,
|
|
limit=50,
|
|
scroll_filter=models.Filter(must=[models.HasVectorCondition(has_vector="multi-code")]),
|
|
)[0],
|
|
)
|