zlei6/ragflow
1
0
Fork 0
mirror of https://github.com/infiniflow/ragflow.git synced 2026-07-20 06:31:02 +08:00
Code Issues Packages Projects Releases Wiki Activity

Refactor: migrate document run api (#14351)

Browse Source 
### What problem does this PR solve?

Before migration: POST /v1/document/run
After migration: POST /api/v1/documents/ingest/

### Type of change

- [x] Refactoring
This commit is contained in:
Jack
2026-04-27 21:25:58 +08:00
committed by GitHub
parent 965717c4fb
commit 49912a156e
9 changed files with 178 additions and 165 deletions
Download Patch File Download Diff File Expand all files Collapse all files

66
api/apps/document_app.py
View File

@@ -20,11 +20,8 @@ from quart import make_response, request
from api.apps import current_user, login_required
from api.constants import IMG_BASE64_PREFIX
from api.db import FileType
from api.db.db_models import Task
from api.db.services.document_service import DocumentService
from api.db.services.file2document_service import File2DocumentService
from api.db.services.knowledgebase_service import KnowledgebaseService
from api.db.services.task_service import TaskService, cancel_all_task_of
from api.utils.api_utils import (
get_data_error_result,
get_json_result,
@@ -58,69 +55,6 @@ def thumbnails():
return server_error_response(e)
@manager.route("/run", methods=["POST"]) # noqa: F821
@login_required
@validate_request("doc_ids", "run")
async def run():
req = await get_request_json()
uid = current_user.id
try:
def _run_sync():
for doc_id in req["doc_ids"]:
if not DocumentService.accessible(doc_id, uid):
return get_json_result(data=False, message="No authorization.", code=RetCode.AUTHENTICATION_ERROR)
kb_table_num_map = {}
for id in req["doc_ids"]:
info = {"run": str(req["run"]), "progress": 0}
if str(req["run"]) == TaskStatus.RUNNING.value and req.get("delete", False):
info["progress_msg"] = ""
info["chunk_num"] = 0
info["token_num"] = 0
tenant_id = DocumentService.get_tenant_id(id)
if not tenant_id:
return get_data_error_result(message="Tenant not found!")
e, doc = DocumentService.get_by_id(id)
if not e:
return get_data_error_result(message="Document not found!")
if str(req["run"]) == TaskStatus.CANCEL.value:
tasks = list(TaskService.query(doc_id=id))
has_unfinished_task = any((task.progress or 0) < 1 for task in tasks)
if str(doc.run) in [TaskStatus.RUNNING.value, TaskStatus.CANCEL.value] or has_unfinished_task:
cancel_all_task_of(id)
else:
return get_data_error_result(message="Cannot cancel a task that is not in RUNNING status")
if all([("delete" not in req or req["delete"]), str(req["run"]) == TaskStatus.RUNNING.value, str(doc.run) == TaskStatus.DONE.value]):
DocumentService.clear_chunk_num_when_rerun(doc.id)
DocumentService.update_by_id(id, info)
if req.get("delete", False):
TaskService.filter_delete([Task.doc_id == id])
if settings.docStoreConn.index_exist(search.index_name(tenant_id), doc.kb_id):
settings.docStoreConn.delete({"doc_id": id}, search.index_name(tenant_id), doc.kb_id)
if str(req["run"]) == TaskStatus.RUNNING.value:
if req.get("apply_kb"):
e, kb = KnowledgebaseService.get_by_id(doc.kb_id)
if not e:
raise LookupError("Can't find this dataset!")
doc.parser_config["llm_id"] = kb.parser_config.get("llm_id")
doc.parser_config["enable_metadata"] = kb.parser_config.get("enable_metadata", False)
doc.parser_config["metadata"] = kb.parser_config.get("metadata", {})
DocumentService.update_parser_config(doc.id, doc.parser_config)
doc_dict = doc.to_dict()
DocumentService.run(tenant_id, doc_dict, kb_table_num_map)
return get_json_result(data=True)
return await thread_pool_exec(_run_sync)
except Exception as e:
return server_error_response(e)
@manager.route("/get/<doc_id>", methods=["GET"]) # noqa: F821
@login_required
async def get(doc_id):