perf: optimize dataset api query speed (#32847)
Some checks failed
autofix.ci / autofix (push) Has been cancelled
Build and Push API & Web / build (api, DIFY_API_IMAGE_NAME, linux/amd64, build-api-amd64) (push) Has been cancelled
Build and Push API & Web / build (api, DIFY_API_IMAGE_NAME, linux/arm64, build-api-arm64) (push) Has been cancelled
Build and Push API & Web / build (web, DIFY_WEB_IMAGE_NAME, linux/amd64, build-web-amd64) (push) Has been cancelled
Build and Push API & Web / build (web, DIFY_WEB_IMAGE_NAME, linux/arm64, build-web-arm64) (push) Has been cancelled
Build and Push API & Web / create-manifest (api, DIFY_API_IMAGE_NAME, merge-api-images) (push) Has been cancelled
Build and Push API & Web / create-manifest (web, DIFY_WEB_IMAGE_NAME, merge-web-images) (push) Has been cancelled
Main CI Pipeline / Check Changed Files (push) Has been cancelled
Main CI Pipeline / API Tests (push) Has been cancelled
Main CI Pipeline / Web Tests (push) Has been cancelled
Main CI Pipeline / Style Check (push) Has been cancelled
Main CI Pipeline / VDB Tests (push) Has been cancelled
Main CI Pipeline / DB Migration Test (push) Has been cancelled

Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com>
This commit is contained in:
wangxiaolei
2026-03-02 18:49:21 +08:00
committed by GitHub
parent c917838f9c
commit e985e73bdc

View File

@@ -53,7 +53,7 @@ from fields.dataset_fields import (
from fields.document_fields import document_status_fields
from libs.login import current_account_with_tenant, login_required
from models import ApiToken, Dataset, Document, DocumentSegment, UploadFile
from models.dataset import DatasetPermissionEnum
from models.dataset import DatasetPermission, DatasetPermissionEnum
from models.provider_ids import ModelProviderID
from services.api_token_service import ApiTokenCache
from services.dataset_service import DatasetPermissionService, DatasetService, DocumentService
@@ -323,6 +323,18 @@ class DatasetListApi(Resource):
model_names.append(f"{embedding_model.model}:{embedding_model.provider.provider}")
data = cast(list[dict[str, Any]], marshal(datasets, dataset_detail_fields))
dataset_ids = [item["id"] for item in data if item.get("permission") == "partial_members"]
partial_members_map: dict[str, list[str]] = {}
if dataset_ids:
permissions = db.session.execute(
select(DatasetPermission.dataset_id, DatasetPermission.account_id).where(
DatasetPermission.dataset_id.in_(dataset_ids)
)
).all()
for dataset_id, account_id in permissions:
partial_members_map.setdefault(dataset_id, []).append(account_id)
for item in data:
# convert embedding_model_provider to plugin standard format
if item["indexing_technique"] == "high_quality" and item["embedding_model_provider"]:
@@ -336,8 +348,7 @@ class DatasetListApi(Resource):
item["embedding_available"] = True
if item.get("permission") == "partial_members":
part_users_list = DatasetPermissionService.get_dataset_partial_member_list(item["id"])
item.update({"partial_member_list": part_users_list})
item.update({"partial_member_list": partial_members_map.get(item["id"], [])})
else:
item.update({"partial_member_list": []})