Files
qdrant/tests/openapi/test_validation.py
Andrey Vasnetsov 8a7325ad5d Remove deprecated search endpoints from OpenAPI, deprecate them in gRPC (#9982)
* Remove deprecated search/recommend/discover endpoints from OpenAPI

Remove deprecated REST API endpoint definitions from the OpenAPI
generator. These endpoints were deprecated in v1.13.3 (`f4ced2567`,
#5907, 2025-01-30) in favor of the universal `/points/query` endpoint:

- POST /points/search
- POST /points/search/batch
- POST /points/search/groups
- POST /points/recommend
- POST /points/recommend/batch
- POST /points/recommend/groups
- POST /points/discover
- POST /points/discover/batch

Also removes the corresponding request types from the schema generator
and updates the expected API count in the consistency check.

Co-authored-by: Cursor <cursoragent@cursor.com>

* Migrate OpenAPI integration tests to /points/query

The deprecated /points/search, /points/recommend and /points/discover
endpoints (along with their /batch and /groups variants) were removed
from the OpenAPI spec, which caused validation failures in the Python
integration test harness.

This commit migrates the affected tests to the universal /points/query
endpoint:

- Delete tests dedicated to the deprecated endpoints:
  test_recommend.py, test_discover.py, test_multicollection_reco.py,
  test_recommendation_multivector.py
- Refactor remaining tests to call /points/query (and /query/batch,
  /query/groups), translating request bodies (vector -> query / using,
  positive/negative -> query.recommend, target/context -> query.discover)
  and unwrapping the new result.points response shape.
- Drop equivalence assertions against the now-removed legacy endpoints.

Co-authored-by: Cursor <cursoragent@cursor.com>

* Relax non-empty assertions in migrated recommend/discover tests

The previous migration added `len(...) > 0` assertions to tests that
previously only checked equivalence between the deprecated and new
API. These assertions are too strict because the parametrized
`query_filter` cases legitimately produce empty result sets.

Drop the `> 0` assertion and rely on `request_with_validation` to
verify the response is well-formed and HTTP OK.

Co-authored-by: Cursor <cursoragent@cursor.com>

* Migrate remaining OpenAPI tests off deprecated search endpoints

Tests added to dev after the original migration was written still call
/points/search and /points/recommend/groups through
`request_with_validation`, which resolves the endpoint against the
OpenAPI spec and therefore breaks once the endpoint is not in the spec:

- test_turbo4_storage.py, test_sparse_idf_corpus.py, test_validation.py:
  translate /points/search to /points/query (vector{name,vector} ->
  query + using, result -> result.points).
- test_group.py: drop the /points/recommend/groups half of the
  lookup_from validation test in favour of the query equivalent.

test_sparse_idf_corpus.py's test_query_api_supports_idf_corpus goes
away: with the helper on /points/query every test in the file now
exercises what it asserted.

Also record why test_recommend_group cannot assert on its groups: it
uses every point in the collection as a recommend example, so all of
them are excluded and the result is legitimately empty.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

* Regenerate openapi.json without the deprecated search endpoints

Drops the 8 deprecated paths and the request schemas that only they
referenced: Search/Recommend/Discover request (+Batch, +Groups) types
and their exclusive dependencies (NamedVector, NamedSparseVector,
NamedVectorStruct, UsingVector, RecommendExample, ContextExamplePair).

Regenerated output is a strict subset of the previous spec, and every
remaining $ref still resolves.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

* Deprecate the search/recommend/discover RPCs in gRPC

The REST counterparts have carried `deprecated: true` since v1.13.3 and
are now gone from the OpenAPI spec, while the gRPC RPCs never got any
deprecation annotation at all. Mark all 8 with `option deprecated = true`
so generated clients warn, and point each doc comment at its `Query`
replacement.

tonic puts `#[deprecated]` on the generated client methods only; the
server trait gets the doc comment alone, so our own `impl` is unaffected.
The RPCs keep serving traffic — this is annotation only.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

* Restore the deleted recommend/discover suites on /points/query

The earlier migration deleted these four files outright, but the
query-side tests it left behind are all shallow smoke tests
(`len(result) > 0`, `"points" in result[0]`). The deleted ones carried
invariants with no query-API equivalent anywhere, so deleting them was a
real loss of coverage rather than de-duplication:

- test_recommend.py: default strategy equals average_vector; batch
  results identical to sequential singles across six request shapes;
  best_score with only negatives yields all-negative scores; best_score
  with a single positive orders identically to a nearest query; raw
  vectors as examples equal ids as examples.
- test_discover.py: context-only scores are all <= 0; target-only orders
  identically to a nearest query but scores differently; with a fixed
  context the integer part of the score is stable while the decimal part
  moves, and vice versa with a fixed target; batch equals singles;
  lookup_from by id equals by vector.
- test_multicollection_reco.py: cross-collection lookup_from, plus
  wrong-vector-size, unknown-collection and unknown-vector rejections.
- test_recommendation_multivector.py: the same recommend invariants over
  a max_sim multivector collection, which the query suite never covered.

Only test_recommend_missing_lookup_from_collection_with_raw_vector is
dropped as genuinely redundant — test_query.py's
test_query_missing_lookup_from_collection covers query, query/batch and
prefetch.

Two request-shape differences the translation had to absorb:

- Giving no examples at all is 422 (a RecommendInput validation rule),
  where the legacy API reported 400 from the query itself. A malformed
  example, such as an empty vector, is still 400.
- DiscoverInput requires the `context` key and accepts only an explicit
  null to mean "no context", so target-only discover must spell it out.
  The legacy API let it be omitted.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

---------

Co-authored-by: Cursor <cursoragent@cursor.com>
Co-authored-by: Claude Opus 5 <noreply@anthropic.com>
2026-07-27 18:09:43 +02:00

231 lines
8.0 KiB
Python

import pytest
import requests
from .helpers.collection_setup import basic_collection_setup, drop_collection
from .helpers.helpers import request_with_validation
from .helpers.settings import QDRANT_HOST
@pytest.fixture(autouse=True, scope="module")
def setup(collection_name):
basic_collection_setup(collection_name=collection_name)
yield
drop_collection(collection_name=collection_name)
def test_validation_collection_name(collection_name):
# Collection names are limited to 255 chars due to filesystem constraints
response = request_with_validation(
api='/collections/{collection_name}/points/{id}',
method="GET",
path_params={
'collection_name': '''\
extremelylongnameextremelylongnameextremelylongname\
extremelylongnameextremelylongnameextremelylongname\
extremelylongnameextremelylongnameextremelylongname\
extremelylongnameextremelylongnameextremelylongname\
extremelylongnameextremelylongnameextremelylongname\
extremelylongnameextremelylongnameextremelylongname\
''',
'id': 1,
},
)
assert not response.ok
assert 'Validation error' in response.json()["status"]["error"]
def test_validation_body_param(collection_name):
# Illegal body parameters must trigger a validation error
response = request_with_validation(
api='/collections/{collection_name}',
method="PUT",
path_params={'collection_name': collection_name},
body={
"vectors": {
"size": 4,
"distance": "Dot"
},
"hnsw_config": {
"ef_construct": 0,
}
}
)
assert not response.ok
assert 'Validation error' in response.json()["status"]["error"]
assert 'hnsw_config.ef_construct' in response.json()["status"]["error"]
def test_validation_search_hnsw_ef_zero(collection_name):
# HNSW search ef must be a positive beam size.
response = request_with_validation(
api='/collections/{collection_name}/points/query',
method="POST",
path_params={'collection_name': collection_name},
body={
"query": [0.2, 0.1, 0.9, 0.7],
"limit": 3,
"params": {"hnsw_ef": 0},
}
)
assert not response.ok
error = response.json()["status"]["error"]
assert 'Validation error' in error
assert 'hnsw_ef' in error
def test_validation_query_param(collection_name):
# Illegal URL parameters must trigger a validation error
response = request_with_validation(
api='/collections/{collection_name}',
method="PUT",
path_params={'collection_name': collection_name},
query_params={'timeout': 0},
body={
"vectors": {
"size": 4,
"distance": "Dot"
},
}
)
assert not response.ok
assert 'Validation error' in response.json()["status"]["error"]
assert 'timeout: value 0 invalid' in response.json()["status"]["error"]
# Regression: two sibling validation errors used to panic
# `common::validation::validate_iter` (validator's internal `add_nested` panics
# on a second insert under the same key), surfacing as a dropped connection
# instead of a 4xx. Length-mismatched sparse vectors with non-empty indices
# fail validation without short-circuiting `VectorStruct::is_empty`.
_INVALID_SPARSE = {"indices": [0, 1], "values": [0.0]}
def test_validation_iter_named_vectors(collection_name):
# Hits VectorStruct::Named -> validate_iter (lib/api/src/rest/schema.rs).
response = request_with_validation(
api='/collections/{collection_name}/points/vectors',
method="PUT",
path_params={'collection_name': collection_name},
query_params={'wait': 'true'},
body={
"points": [
{
"id": 1,
"vector": {"a": _INVALID_SPARSE, "b": _INVALID_SPARSE},
}
],
},
)
assert not response.ok
assert 'Validation error' in response.json()["status"]["error"]
def test_validation_iter_batch_named_vectors(collection_name):
# Hits BatchVectorStruct::Named -> validate_iter (lib/api/src/rest/validate.rs).
response = request_with_validation(
api='/collections/{collection_name}/points/batch',
method="POST",
path_params={'collection_name': collection_name},
query_params={'wait': 'true'},
body={
"operations": [
{
"upsert": {
"batch": {
"ids": [1, 2],
"vectors": {"a": [_INVALID_SPARSE, _INVALID_SPARSE]},
}
}
}
]
},
)
assert not response.ok
assert 'Validation error' in response.json()["status"]["error"]
# Regression for https://github.com/qdrant/qdrant/issues/9045
#
# Upserting an empty vector `[]` is rejected on the synchronous (`wait=true`)
# path but silently accepted on the asynchronous path: the response is HTTP 200
# `acknowledged`, the point is later discarded, and the zero-length vector can
# reach internal code paths that assert on non-zero length (see #7967).
@pytest.mark.parametrize("wait", ["true", "false"])
def test_validation_empty_vector_upsert(collection_name, wait):
response = request_with_validation(
api='/collections/{collection_name}/points',
method="PUT",
path_params={'collection_name': collection_name},
query_params={'wait': wait},
body={"points": [{"id": 1000, "vector": []}]},
)
assert not response.ok, (
f"empty vector accepted with wait={wait}: "
f"status={response.status_code}, body={response.text}"
)
@pytest.mark.parametrize("wait", ["true", "false"])
def test_validation_empty_vector_batch_upsert(collection_name, wait):
response = request_with_validation(
api='/collections/{collection_name}/points/batch',
method="POST",
path_params={'collection_name': collection_name},
query_params={'wait': wait},
body={
"operations": [
{"upsert": {"points": [{"id": 1001, "vector": []}]}},
],
},
)
assert not response.ok, (
f"empty vector accepted in batch upsert with wait={wait}: "
f"status={response.status_code}, body={response.text}"
)
# Regression for https://github.com/qdrant/qdrant/issues/9149
#
# `shard_number`, `replication_factor`, and `write_consistency_factor` must be
# at least 1. Bypasses `request_with_validation` (which short-circuits on the
# client-side OpenAPI minimum) so the server-side `Validate` derive is what
# actually rejects the request — that's the contract we care about.
@pytest.mark.parametrize(
"field",
["shard_number", "replication_factor", "write_consistency_factor"],
)
def test_validation_positive_integer_zero(field):
name = f"test_validation_{field}_zero"
response = requests.put(
f"{QDRANT_HOST}/collections/{name}",
json={
"vectors": {"size": 4, "distance": "Dot"},
field: 0,
},
)
assert response.status_code == 422, (
f"expected 422 for {field}=0, got {response.status_code}: {response.text}"
)
assert field in response.json()["status"]["error"]
# Negative integers can't fit in `u32`, so serde rejects them at deserialization
# (HTTP 400) before the `Validate` derive runs — different status, same outcome.
@pytest.mark.parametrize(
"field",
["shard_number", "replication_factor", "write_consistency_factor"],
)
def test_validation_positive_integer_negative(field):
name = f"test_validation_{field}_neg"
response = requests.put(
f"{QDRANT_HOST}/collections/{name}",
json={
"vectors": {"size": 4, "distance": "Dot"},
field: -1,
},
)
assert response.status_code == 400, (
f"expected 400 for {field}=-1, got {response.status_code}: {response.text}"
)