AI/ragflow
1
0
Fork 0
mirror of https://git.mirrors.martin98.com/https://github.com/infiniflow/ragflow.git synced 2025-08-14 06:26:00 +08:00
Code Issues Packages Projects Releases Wiki Activity

Fix out of boundary. (#3786)

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

#3769
### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)
This commit is contained in:
Kevin Hu 2024-12-02 11:38:53 +08:00 committed by GitHub
parent a7efd3cac5
commit 7058ac0041
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
2 changed files with 25 additions and 23 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

44
api/apps/sdk/session.py
View File

@ -35,7 +35,7 @@ from api.db.services.llm_service import LLMBundle
@manager.route('/chats/<chat_id>/sessions', methods=['POST']) @manager.route('/chats/<chat_id>/sessions', methods=['POST'])
@token_required @token_required
def create(tenant_id,chat_id): def create(tenant_id, chat_id):
req = request.json req = request.json
req["dialog_id"] = chat_id req["dialog_id"] = chat_id
dia = DialogService.query(tenant_id=tenant_id, id=req["dialog_id"], status=StatusEnum.VALID.value) dia = DialogService.query(tenant_id=tenant_id, id=req["dialog_id"], status=StatusEnum.VALID.value)
@ -77,7 +77,7 @@ def create_agent_session(tenant_id, agent_id):
conv = { conv = {
"id": get_uuid(), "id": get_uuid(),
"dialog_id": cvs.id, "dialog_id": cvs.id,
"user_id": req.get("usr_id","") if isinstance(req, dict) else "", "user_id": req.get("usr_id", "") if isinstance(req, dict) else "",
"message": [{"role": "assistant", "content": canvas.get_prologue()}], "message": [{"role": "assistant", "content": canvas.get_prologue()}],
"source": "agent" "source": "agent"
} }
@ -88,11 +88,11 @@ def create_agent_session(tenant_id, agent_id):
@manager.route('/chats/<chat_id>/sessions/<session_id>', methods=['PUT']) @manager.route('/chats/<chat_id>/sessions/<session_id>', methods=['PUT'])
@token_required @token_required
def update(tenant_id,chat_id,session_id): def update(tenant_id, chat_id, session_id):
req = request.json req = request.json
req["dialog_id"] = chat_id req["dialog_id"] = chat_id
conv_id = session_id conv_id = session_id
conv = ConversationService.query(id=conv_id,dialog_id=chat_id) conv = ConversationService.query(id=conv_id, dialog_id=chat_id)
if not conv: if not conv:
return get_error_data_result(message="Session does not exist") return get_error_data_result(message="Session does not exist")
if not DialogService.query(id=chat_id, tenant_id=tenant_id, status=StatusEnum.VALID.value): if not DialogService.query(id=chat_id, tenant_id=tenant_id, status=StatusEnum.VALID.value):
@ -123,12 +123,12 @@ def completion(tenant_id, chat_id):
return get_error_data_result(message="`name` can not be empty.") return get_error_data_result(message="`name` can not be empty.")
ConversationService.save(**conv) ConversationService.save(**conv)
e, conv = ConversationService.get_by_id(conv["id"]) e, conv = ConversationService.get_by_id(conv["id"])
session_id=conv.id session_id = conv.id
else: else:
session_id = req.get("session_id") session_id = req.get("session_id")
if not req.get("question"): if not req.get("question"):
return get_error_data_result(message="Please input your question.") return get_error_data_result(message="Please input your question.")
conv = ConversationService.query(id=session_id,dialog_id=chat_id) conv = ConversationService.query(id=session_id, dialog_id=chat_id)
if not conv: if not conv:
return get_error_data_result(message="Session does not exist") return get_error_data_result(message="Session does not exist")
conv = conv[0] conv = conv[0]
@ -182,7 +182,7 @@ def completion(tenant_id, chat_id):
chunk_list.append(new_chunk) chunk_list.append(new_chunk)
reference["chunks"] = chunk_list reference["chunks"] = chunk_list
ans["id"] = message_id ans["id"] = message_id
ans["session_id"]=session_id ans["session_id"] = session_id
def stream(): def stream():
nonlocal dia, msg, req, conv nonlocal dia, msg, req, conv
@ -193,7 +193,7 @@ def completion(tenant_id, chat_id):
ConversationService.update_by_id(conv.id, conv.to_dict()) ConversationService.update_by_id(conv.id, conv.to_dict())
except Exception as e: except Exception as e:
yield "data:" + json.dumps({"code": 500, "message": str(e), yield "data:" + json.dumps({"code": 500, "message": str(e),
"data": {"answer": "**ERROR**: " + str(e),"reference": []}}, "data": {"answer": "**ERROR**: " + str(e), "reference": []}},
ensure_ascii=False) + "\n\n" ensure_ascii=False) + "\n\n"
yield "data:" + json.dumps({"code": 0, "data": True}, ensure_ascii=False) + "\n\n" yield "data:" + json.dumps({"code": 0, "data": True}, ensure_ascii=False) + "\n\n"
@ -235,7 +235,7 @@ def agent_completion(tenant_id, agent_id):
conv = { conv = {
"id": session_id, "id": session_id,
"dialog_id": cvs.id, "dialog_id": cvs.id,
"user_id": req.get("user_id",""), "user_id": req.get("user_id", ""),
"message": [{"role": "assistant", "content": canvas.get_prologue()}], "message": [{"role": "assistant", "content": canvas.get_prologue()}],
"source": "agent" "source": "agent"
} }
@ -251,9 +251,9 @@ def agent_completion(tenant_id, agent_id):
question = req.get("question") question = req.get("question")
if not question: if not question:
return get_error_data_result("`question` is required.") return get_error_data_result("`question` is required.")
question={ question = {
"role":"user", "role": "user",
"content":question, "content": question,
"id": str(uuid4()) "id": str(uuid4())
} }
messages.append(question) messages.append(question)
@ -375,7 +375,7 @@ def agent_completion(tenant_id, agent_id):
@manager.route('/chats/<chat_id>/sessions', methods=['GET']) @manager.route('/chats/<chat_id>/sessions', methods=['GET'])
@token_required @token_required
def list_session(chat_id,tenant_id): def list_session(chat_id, tenant_id):
if not DialogService.query(tenant_id=tenant_id, id=chat_id, status=StatusEnum.VALID.value): if not DialogService.query(tenant_id=tenant_id, id=chat_id, status=StatusEnum.VALID.value):
return get_error_data_result(message=f"You don't own the assistant {chat_id}.") return get_error_data_result(message=f"You don't own the assistant {chat_id}.")
id = request.args.get("id") id = request.args.get("id")
@ -387,7 +387,7 @@ def list_session(chat_id,tenant_id):
desc = False desc = False
else: else:
desc = True desc = True
convs = ConversationService.get_list(chat_id,page_number,items_per_page,orderby,desc,id,name) convs = ConversationService.get_list(chat_id, page_number, items_per_page, orderby, desc, id, name)
if not convs: if not convs:
return get_result(data=[]) return get_result(data=[])
for conv in convs: for conv in convs:
@ -429,7 +429,7 @@ def list_session(chat_id,tenant_id):
@manager.route('/chats/<chat_id>/sessions', methods=["DELETE"]) @manager.route('/chats/<chat_id>/sessions', methods=["DELETE"])
@token_required @token_required
def delete(tenant_id,chat_id): def delete(tenant_id, chat_id):
if not DialogService.query(id=chat_id, tenant_id=tenant_id, status=StatusEnum.VALID.value): if not DialogService.query(id=chat_id, tenant_id=tenant_id, status=StatusEnum.VALID.value):
return get_error_data_result(message="You don't own the chat") return get_error_data_result(message="You don't own the chat")
req = request.json req = request.json
@ -437,21 +437,22 @@ def delete(tenant_id,chat_id):
if not req: if not req:
ids = None ids = None
else: else:
ids=req.get("ids") ids = req.get("ids")
if not ids: if not ids:
conv_list = [] conv_list = []
for conv in convs: for conv in convs:
conv_list.append(conv.id) conv_list.append(conv.id)
else: else:
conv_list=ids conv_list = ids
for id in conv_list: for id in conv_list:
conv = ConversationService.query(id=id,dialog_id=chat_id) conv = ConversationService.query(id=id, dialog_id=chat_id)
if not conv: if not conv:
return get_error_data_result(message="The chat doesn't own the session") return get_error_data_result(message="The chat doesn't own the session")
ConversationService.delete_by_id(id) ConversationService.delete_by_id(id)
return get_result() return get_result()
@manager.route('/sessions/ask', methods=['POST']) @manager.route('/sessions/ask', methods=['POST'])
@token_required @token_required
def ask_about(tenant_id): def ask_about(tenant_id):
@ -460,17 +461,18 @@ def ask_about(tenant_id):
return get_error_data_result("`question` is required.") return get_error_data_result("`question` is required.")
if not req.get("dataset_ids"): if not req.get("dataset_ids"):
return get_error_data_result("`dataset_ids` is required.") return get_error_data_result("`dataset_ids` is required.")
if not isinstance(req.get("dataset_ids"),list): if not isinstance(req.get("dataset_ids"), list):
return get_error_data_result("`dataset_ids` should be a list.") return get_error_data_result("`dataset_ids` should be a list.")
req["kb_ids"]=req.pop("dataset_ids") req["kb_ids"] = req.pop("dataset_ids")
for kb_id in req["kb_ids"]: for kb_id in req["kb_ids"]:
if not KnowledgebaseService.accessible(kb_id,tenant_id): if not KnowledgebaseService.accessible(kb_id, tenant_id):
return get_error_data_result(f"You don't own the dataset {kb_id}.") return get_error_data_result(f"You don't own the dataset {kb_id}.")
kbs = KnowledgebaseService.query(id=kb_id) kbs = KnowledgebaseService.query(id=kb_id)
kb = kbs[0] kb = kbs[0]
if kb.chunk_num == 0: if kb.chunk_num == 0:
return get_error_data_result(f"The dataset {kb_id} doesn't own parsed file") return get_error_data_result(f"The dataset {kb_id} doesn't own parsed file")
uid = tenant_id uid = tenant_id
def stream(): def stream():
nonlocal req, uid nonlocal req, uid
try: try:

2
deepdoc/parser/pdf_parser.py
View File

@ -152,7 +152,7 @@ class RAGFlowPdfParser:
max(len(up["text"]), len(down["text"])), max(len(up["text"]), len(down["text"])),
len(tks_all) - len(tks_up) - len(tks_down), len(tks_all) - len(tks_up) - len(tks_down),
len(tks_down) - len(tks_up), len(tks_down) - len(tks_up),
tks_down[-1] == tks_up[-1], tks_down[-1] == tks_up[-1] if tks_down and tks_up else False,
max(down["in_row"], up["in_row"]), max(down["in_row"], up["in_row"]),
abs(down["in_row"] - up["in_row"]), abs(down["in_row"] - up["in_row"]),
len(tks_down) == 1 and rag_tokenizer.tag(tks_down[0]).find("n") >= 0, len(tks_down) == 1 and rag_tokenizer.tag(tks_down[0]).find("n") >= 0,