RAGFlow is an agent-native database: a Retrieval-Augmented Generation (RAG) engine that grew into context infrastructure.
Go agent-harnessagent-native-databaseagentic-aiagentic-retrievalagentic-searchaiai-agentscontext-enginecontext-engineeringcontext-managementharness-engineeringknowledge-compilationragretrieval-augmented-generationretrieval-harnesssearch-harness
feat: sort sync log list by status priority (#19777)
mkaaad36a6456f4df5
diff --git a/api/db/services/connector_service.py b/api/db/services/connector_service.pyindex a44570fabb..a108ac386f 100644--- a/api/db/services/connector_service.py+++ b/api/db/services/connector_service.py@@ -14,23 +14,21 @@# limitations under the License.#import logging-from datetime import datetimeimport os-from typing import Optional, Tuple, List+from datetime import datetimefrom anthropic import BaseModel-from peewee import SQL, fn+from peewee import SQL, Case, fnfrom api.db import InputType-from api.db.db_models import DB, Connector, SyncLogs, Connector2Kb, Knowledgebase+from api.db.db_models import DB, Connector, Connector2Kb, Knowledgebase, SyncLogsfrom api.db.services.common_service import CommonService-from api.db.services.document_service import DocumentService-from api.db.services.document_service import DocMetadataService+from api.db.services.document_service import DocMetadataService, DocumentServicefrom api.db.services.knowledgebase_service import KnowledgebaseServicefrom api.utils.common import hash128from common import settings-from common.misc_utils import get_uuidfrom common.constants import ConnectorTaskType, TaskStatus+from common.misc_utils import get_uuidfrom common.settings import TIMEZONEfrom common.time_utils import current_timestamp, timestamp_to_date@@ -302,7 +300,7 @@ class SyncLogsService(CommonService):model = SyncLogs@classmethod- def list_sync_tasks(cls, connector_id=None, page_number=None, items_per_page=15) -> Tuple[List[dict], int]:+ def list_sync_tasks(cls, connector_id=None, page_number=None, items_per_page=15) -> tuple[list[dict], int]:fields = [cls.model.id,cls.model.connector_id,@@ -320,6 +318,14 @@ class SyncLogsService(CommonService):Knowledgebase.name.alias("kb_name"),cls.model.status,cls.model.update_time,+ Case(+ None,+ [+ (cls.model.status == TaskStatus.SCHEDULE, 0),+ (cls.model.status == TaskStatus.RUNNING, 1),+ ],+ 2,+ ).alias("status_rank"),]if not connector_id:fields.append(Connector.config)@@ -344,22 +350,25 @@ class SyncLogsService(CommonService):expr = SQL("NOW() - INTERVAL `t2`.`refresh_freq` MINUTE")query = query.where(Connector.input_type == InputType.POLL, Connector.status == TaskStatus.SCHEDULE, cls.model.status == TaskStatus.SCHEDULE, cls.model.update_date < expr)- query = query.distinct().order_by(cls.model.update_time.desc())+ query = query.distinct().order_by(SQL("status_rank"), cls.model.update_time.desc())total = query.count()if page_number:query = query.paginate(page_number, items_per_page)- return list(query.dicts()), total+ rows = list(query.dicts())+ for row in rows:+ row.pop("status_rank", None)+ return rows, total@classmethod- def list_due_sync_tasks(cls) -> List[dict]:+ def list_due_sync_tasks(cls) -> list[dict]:return cls._list_due_tasks_for_freq(ConnectorTaskType.SYNC,"refresh_freq",)@classmethod- def list_due_prune_tasks(cls) -> List[dict]:+ def list_due_prune_tasks(cls) -> list[dict]:tasks = cls._list_due_tasks_for_freq(ConnectorTaskType.PRUNE,"prune_freq",@@ -373,7 +382,7 @@ class SyncLogsService(CommonService):]@classmethod- def _list_due_tasks_for_freq(cls, task_type: str, freq_field: str) -> List[dict]:+ def _list_due_tasks_for_freq(cls, task_type: str, freq_field: str) -> list[dict]:fields = [cls.model.id,cls.model.connector_id,@@ -476,17 +485,15 @@ class SyncLogsService(CommonService):task_id = get_uuid()ConnectorService.update_by_id(connector_id, {"status": TaskStatus.SCHEDULE})ret = cls.save(- **{- "id": task_id,- "kb_id": kb_id,- "status": TaskStatus.SCHEDULE,- "connector_id": connector_id,- "task_type": task_type,- "poll_range_start": poll_range_start,- "from_beginning": reindex,- "total_docs_indexed": total_docs_indexed,- "time_started": datetime.now().strftime("%Y-%m-%d %H:%M:%S"),- }+ id=task_id,+ kb_id=kb_id,+ status=TaskStatus.SCHEDULE,+ connector_id=connector_id,+ task_type=task_type,+ poll_range_start=poll_range_start,+ from_beginning=reindex,+ total_docs_indexed=total_docs_indexed,+ time_started=datetime.now().strftime("%Y-%m-%d %H:%M:%S"),)if run_immediately:DB.execute_sql(@@ -554,7 +561,7 @@ class SyncLogsService(CommonService):id: strfilename: strblob: bytes- fingerprint: Optional[str] = None+ fingerprint: str | None = Nonedef read(self) -> bytes:return self.blob@@ -624,7 +631,7 @@ class Connector2KbService(CommonService):if conn_id in old_conn_ids:cls.filter_update([cls.model.connector_id == conn_id, cls.model.kb_id == kb_id], {"auto_parse": conn.get("auto_parse", "1")})continue- cls.save(**{"id": get_uuid(), "connector_id": conn_id, "kb_id": kb_id, "auto_parse": conn.get("auto_parse", "1")})+ cls.save(id=get_uuid(), connector_id=conn_id, kb_id=kb_id, auto_parse=conn.get("auto_parse", "1"))SyncLogsService.schedule(conn_id, kb_id, reindex=True, task_type=ConnectorTaskType.SYNC, run_immediately=True)e, full_conn = ConnectorService.get_by_id(conn_id)if e and (full_conn.config or {}).get("sync_deleted_files"):diff --git a/internal/dao/connector.go b/internal/dao/connector.goindex 8ec0b626b1..1255eb73b1 100644--- a/internal/dao/connector.go+++ b/internal/dao/connector.go@@ -516,6 +516,18 @@ func connectorConfigBool(config map[string]interface{}, key string) bool {}}+// syncLogsStatusPriorityOrder orders sync log rows by status priority:+// SCHEDULE first, RUNNING second, then the rest, newest update first within+// each group. Returned as a single raw ORDER BY fragment because gorm.Expr is+// silently dropped by gorm's Order and repeated Order calls merge in reverse.+func syncLogsStatusPriorityOrder() string {+ return fmt.Sprintf(+ "CASE WHEN sync_logs.status = '%s' THEN 0 WHEN sync_logs.status = '%s' THEN 1 ELSE 2 END, sync_logs.update_time DESC",+ string(entity.TaskStatusSchedule),+ string(entity.TaskStatusRunning),+ )+}+// ListLogsByConnectorID lists sync logs for one connector with pagination.func (dao *ConnectorDAO) ListLogsByConnectorID(ctx context.Context, db *gorm.DB, connectorID string, offset, limit int) ([]*entity.ConnectorSyncLog, int64, error) {baseQuery := db.WithContext(ctx).Model(&entity.SyncLogs{}).@@ -547,9 +559,10 @@ func (dao *ConnectorDAO) ListLogsByConnectorID(ctx context.Context, db *gorm.DB,"connector.prune_freq AS prune_freq","knowledgebase.name AS kb_name","sync_logs.status",+ "sync_logs.update_time",).Distinct().- Order("sync_logs.update_date DESC").+ Order(syncLogsStatusPriorityOrder()).Offset(offset).Limit(limit).Scan(&logs).Error@@ -596,9 +609,10 @@ func (dao *ConnectorDAO) ListLogs(ctx context.Context, db *gorm.DB, tenantIDs []"connector.prune_freq AS prune_freq","knowledgebase.name AS kb_name","sync_logs.status",+ "sync_logs.update_time",).Distinct().- Order("sync_logs.update_date DESC")+ Order(syncLogsStatusPriorityOrder())if limit > 0 {query = query.Offset(offset).Limit(limit)}diff --git a/internal/service/connector_logs_test.go b/internal/service/connector_logs_test.goindex 308aa32380..e5e8e9ed7c 100644--- a/internal/service/connector_logs_test.go+++ b/internal/service/connector_logs_test.go@@ -20,6 +20,7 @@ import ("context""fmt""testing"+ "time""ragflow/internal/common""ragflow/internal/dao"@@ -112,3 +113,104 @@ func TestListLogsPaginationModes(t *testing.T) {t.Fatalf("ListLogs(normalize) = total %d len %d, want 5/5", total, len(normalized))}}++func i64ptr(v int64) *int64 { return &v }+func timeMillisPtr(v int64) *time.Time { ts := time.UnixMilli(v); return &ts }++func TestListLogsStatusPriority(t *testing.T) {+ db := setupServiceTestDB(t)+ pushServiceDB(t, db)+ if err := db.AutoMigrate(&entity.Connector{}, &entity.Connector2Kb{}, &entity.Knowledgebase{}, &entity.SyncLogs{}); err != nil {+ t.Fatalf("migrate connector tables: %v", err)+ }++ if err := db.Create(&entity.Connector{+ ID: "conn-1",+ TenantID: "user-1",+ Name: "conn-1",+ Source: "rss",+ InputType: "poll",+ Config: entity.JSONMap{},+ Status: string(entity.TaskStatusSchedule),+ RefreshFreq: 5,+ PruneFreq: 5,+ TimeoutSecs: 60,+ }).Error; err != nil {+ t.Fatalf("insert connector: %v", err)+ }+ if err := db.Create(&entity.Knowledgebase{+ ID: "kb-1",+ TenantID: "user-1",+ Name: "kb-1",+ CreatedBy: "user-1",+ EmbdID: "embd",+ }).Error; err != nil {+ t.Fatalf("insert kb: %v", err)+ }+ if err := db.Create(&entity.Connector2Kb{+ ID: "conn-1-kb-1",+ ConnectorID: "conn-1",+ KbID: "kb-1",+ AutoParse: "1",+ }).Error; err != nil {+ t.Fatalf("insert connector2kb: %v", err)+ }++ // Deliberately interleave statuses and update_time so a missing status+ // priority or a missing secondary sort is observable in the result order.+ seed := []entity.SyncLogs{+ {ID: "task-done-1", ConnectorID: "conn-1", KbID: "kb-1", TaskType: dao.TaskTypeSync, Status: string(entity.TaskStatusDone), BaseModel: entity.BaseModel{UpdateTime: i64ptr(100), UpdateDate: timeMillisPtr(100)}},+ {ID: "task-done-2", ConnectorID: "conn-1", KbID: "kb-1", TaskType: dao.TaskTypeSync, Status: string(entity.TaskStatusDone), BaseModel: entity.BaseModel{UpdateTime: i64ptr(200), UpdateDate: timeMillisPtr(200)}},+ {ID: "task-running-1", ConnectorID: "conn-1", KbID: "kb-1", TaskType: dao.TaskTypeSync, Status: string(entity.TaskStatusRunning), BaseModel: entity.BaseModel{UpdateTime: i64ptr(300), UpdateDate: timeMillisPtr(300)}},+ {ID: "task-schedule-1", ConnectorID: "conn-1", KbID: "kb-1", TaskType: dao.TaskTypeSync, Status: string(entity.TaskStatusSchedule), BaseModel: entity.BaseModel{UpdateTime: i64ptr(400), UpdateDate: timeMillisPtr(400)}},+ {ID: "task-fail-1", ConnectorID: "conn-1", KbID: "kb-1", TaskType: dao.TaskTypeSync, Status: string(entity.TaskStatusFail), BaseModel: entity.BaseModel{UpdateTime: i64ptr(500), UpdateDate: timeMillisPtr(500)}},+ {ID: "task-schedule-2", ConnectorID: "conn-1", KbID: "kb-1", TaskType: dao.TaskTypeSync, Status: string(entity.TaskStatusSchedule), BaseModel: entity.BaseModel{UpdateTime: i64ptr(600), UpdateDate: timeMillisPtr(600)}},+ }+ for _, log := range seed {+ if err := db.Create(&log).Error; err != nil {+ t.Fatalf("insert sync log %s: %v", log.ID, err)+ }+ }++ want := []string{+ "task-schedule-2",+ "task-schedule-1",+ "task-running-1",+ "task-fail-1",+ "task-done-2",+ "task-done-1",+ }++ svc := NewConnectorService()+ ctx := context.Background()++ byConnector, total, code, err := svc.ListLog(ctx, "conn-1", "user-1", 1, 100)+ if err != nil || code != common.CodeSuccess {+ t.Fatalf("ListLog: code=%v err=%v", code, err)+ }+ if total != int64(len(want)) {+ t.Fatalf("ListLog total = %d, want %d", total, len(want))+ }+ assertLogOrder(t, "ListLog", byConnector, want)++ tenantWide, total, code, err := svc.ListLogs(ctx, "user-1", "", 1, 100)+ if err != nil || code != common.CodeSuccess {+ t.Fatalf("ListLogs: code=%v err=%v", code, err)+ }+ if total != int64(len(want)) {+ t.Fatalf("ListLogs total = %d, want %d", total, len(want))+ }+ assertLogOrder(t, "ListLogs", tenantWide, want)+}++func assertLogOrder(t *testing.T, name string, logs []*entity.ConnectorSyncLog, want []string) {+ t.Helper()+ if len(logs) != len(want) {+ t.Fatalf("%s returned %d logs, want %d", name, len(logs), len(want))+ }+ for i, log := range logs {+ if log.ID != want[i] {+ t.Fatalf("%s order[%d] = %s, want %s", name, i, log.ID, want[i])+ }+ }+}diff --git a/test/unit_test/api/db/services/test_connector_sync_logs_order.py b/test/unit_test/api/db/services/test_connector_sync_logs_order.pynew file mode 100644index 0000000000..fe67c8a16a--- /dev/null+++ b/test/unit_test/api/db/services/test_connector_sync_logs_order.py@@ -0,0 +1,78 @@+#+# Copyright 2026 The InfiniFlow Authors. All Rights Reserved.+#+# Licensed under the Apache License, Version 2.0 (the "License");+# you may not use this file except in compliance with the License.+# You may obtain a copy of the License at+#+# http://www.apache.org/licenses/LICENSE-2.0+#+# Unless required by applicable law or agreed to in writing, software+# distributed under the License is distributed on an "AS IS" BASIS,+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.+# See the License for the specific language governing permissions and+# limitations under the License.+#+"""Sync task list orders SCHEDULE first, RUNNING second, then the rest."""++from datetime import UTC, datetime++from peewee import SqliteDatabase++from api.db.db_models import DB, Connector, Connector2Kb, Knowledgebase, SyncLogs+from api.db.services.connector_service import SyncLogsService+from common.constants import TaskStatus++_MODELS = [SyncLogs, Connector, Connector2Kb, Knowledgebase]+++def test_list_sync_tasks_status_priority():+ sqlite_db = SqliteDatabase(":memory:")+ sqlite_db.bind(_MODELS, bind_refs=False, bind_backrefs=False)+ sqlite_db.connect()+ sqlite_db.create_tables(_MODELS)+ try:+ Connector.create(+ id="conn-1",+ tenant_id="tenant-1",+ name="conn-1",+ source="rss",+ input_type="poll",+ status=TaskStatus.SCHEDULE,+ refresh_freq=5,+ prune_freq=5,+ timeout_secs=60,+ )+ Knowledgebase.create(id="kb-1", tenant_id="tenant-1", name="kb-1", embd_id="", created_by="tenant-1")+ Connector2Kb.create(id="c2k-1", connector_id="conn-1", kb_id="kb-1", auto_parse="1")++ rows = [+ ("task-done-1", TaskStatus.DONE, 100),+ ("task-done-2", TaskStatus.DONE, 200),+ ("task-running-1", TaskStatus.RUNNING, 300),+ ("task-schedule-1", TaskStatus.SCHEDULE, 400),+ ("task-fail-1", TaskStatus.FAIL, 500),+ ("task-schedule-2", TaskStatus.SCHEDULE, 600),+ ]+ for task_id, status, ts in rows:+ SyncLogs.create(id=task_id, connector_id="conn-1", kb_id="kb-1", status=status)+ sqlite_db.execute_sql(+ "UPDATE sync_logs SET update_time = ?, update_date = ? WHERE id = ?",+ (ts, datetime.fromtimestamp(ts / 1000, tz=UTC), task_id),+ )++ logs, total = SyncLogsService.list_sync_tasks("conn-1", 1, 100)+ assert total == len(rows)+ assert [log["id"] for log in logs] == [+ "task-schedule-2",+ "task-schedule-1",+ "task-running-1",+ "task-fail-1",+ "task-done-2",+ "task-done-1",+ ]+ assert all("status_rank" not in log for log in logs)+ finally:+ sqlite_db.drop_tables(_MODELS)+ sqlite_db.close()+ DB.bind(_MODELS, bind_refs=False, bind_backrefs=False)