mirror of
https://github.com/infiniflow/ragflow.git
synced 2025-12-08 20:42:30 +08:00
Add Q&A and Book, fix task running bugs (#50)
This commit is contained in:
@ -18,10 +18,12 @@ import datetime
|
||||
from flask import request
|
||||
from flask_login import login_required, current_user
|
||||
from elasticsearch_dsl import Q
|
||||
|
||||
from rag.app.qa import rmPrefix, beAdoc
|
||||
from rag.nlp import search, huqie, retrievaler
|
||||
from rag.utils import ELASTICSEARCH, rmSpace
|
||||
from api.db import LLMType
|
||||
from api.db.services.kb_service import KnowledgebaseService
|
||||
from api.db import LLMType, ParserType
|
||||
from api.db.services.knowledgebase_service import KnowledgebaseService
|
||||
from api.db.services.llm_service import TenantLLMService
|
||||
from api.db.services.user_service import UserTenantService
|
||||
from api.utils.api_utils import server_error_response, get_data_error_result, validate_request
|
||||
@ -89,10 +91,8 @@ def get():
|
||||
res["chunk_id"] = id
|
||||
k = []
|
||||
for n in res.keys():
|
||||
if re.search(r"(_vec$|_sm_)", n):
|
||||
if re.search(r"(_vec$|_sm_|_tks|_ltks)", n):
|
||||
k.append(n)
|
||||
if re.search(r"(_tks|_ltks)", n):
|
||||
res[n] = rmSpace(res[n])
|
||||
for n in k:
|
||||
del res[n]
|
||||
|
||||
@ -106,12 +106,12 @@ def get():
|
||||
|
||||
@manager.route('/set', methods=['POST'])
|
||||
@login_required
|
||||
@validate_request("doc_id", "chunk_id", "content_ltks",
|
||||
@validate_request("doc_id", "chunk_id", "content_with_weight",
|
||||
"important_kwd")
|
||||
def set():
|
||||
req = request.json
|
||||
d = {"id": req["chunk_id"]}
|
||||
d["content_ltks"] = huqie.qie(req["content_ltks"])
|
||||
d["content_ltks"] = huqie.qie(req["content_with_weight"])
|
||||
d["content_sm_ltks"] = huqie.qieqie(d["content_ltks"])
|
||||
d["important_kwd"] = req["important_kwd"]
|
||||
d["important_tks"] = huqie.qie(" ".join(req["important_kwd"]))
|
||||
@ -127,8 +127,15 @@ def set():
|
||||
e, doc = DocumentService.get_by_id(req["doc_id"])
|
||||
if not e:
|
||||
return get_data_error_result(retmsg="Document not found!")
|
||||
|
||||
if doc.parser_id == ParserType.QA:
|
||||
arr = [t for t in re.split(r"[\n\t]", req["content_with_weight"]) if len(t)>1]
|
||||
if len(arr) != 2: return get_data_error_result(retmsg="Q&A must be separated by TAB/ENTER key.")
|
||||
q, a = rmPrefix(arr[0]), rmPrefix[arr[1]]
|
||||
d = beAdoc(d, arr[0], arr[1], not any([huqie.is_chinese(t) for t in q+a]))
|
||||
|
||||
v, c = embd_mdl.encode([doc.name, req["content_ltks"]])
|
||||
v = 0.1 * v[0] + 0.9 * v[1]
|
||||
v = 0.1 * v[0] + 0.9 * v[1] if doc.parser_id != ParserType.QA else v[1]
|
||||
d["q_%d_vec" % len(v)] = v.tolist()
|
||||
ELASTICSEARCH.upsert([d], search.index_name(tenant_id))
|
||||
return get_json_result(data=True)
|
||||
|
||||
@ -18,7 +18,7 @@ from flask import request
|
||||
from flask_login import login_required, current_user
|
||||
from api.db.services.dialog_service import DialogService
|
||||
from api.db import StatusEnum
|
||||
from api.db.services.kb_service import KnowledgebaseService
|
||||
from api.db.services.knowledgebase_service import KnowledgebaseService
|
||||
from api.db.services.user_service import TenantService
|
||||
from api.utils.api_utils import server_error_response, get_data_error_result, validate_request
|
||||
from api.utils import get_uuid
|
||||
|
||||
@ -27,10 +27,10 @@ from api.db.services.task_service import TaskService
|
||||
from rag.nlp import search
|
||||
from rag.utils import ELASTICSEARCH
|
||||
from api.db.services import duplicate_name
|
||||
from api.db.services.kb_service import KnowledgebaseService
|
||||
from api.db.services.knowledgebase_service import KnowledgebaseService
|
||||
from api.utils.api_utils import server_error_response, get_data_error_result, validate_request
|
||||
from api.utils import get_uuid
|
||||
from api.db import FileType
|
||||
from api.db import FileType, TaskStatus
|
||||
from api.db.services.document_service import DocumentService
|
||||
from api.settings import RetCode
|
||||
from api.utils.api_utils import get_json_result
|
||||
@ -210,13 +210,12 @@ def rm():
|
||||
@manager.route('/run', methods=['POST'])
|
||||
@login_required
|
||||
@validate_request("doc_ids", "run")
|
||||
def rm():
|
||||
def run():
|
||||
req = request.json
|
||||
try:
|
||||
for id in req["doc_ids"]:
|
||||
DocumentService.update_by_id(id, {"run": str(req["run"])})
|
||||
if req["run"] == "2":
|
||||
TaskService.filter_delete([Task.doc_id == id])
|
||||
DocumentService.update_by_id(id, {"run": str(req["run"]), "progress": 0})
|
||||
if str(req["run"]) == TaskStatus.CANCEL.value:
|
||||
tenant_id = DocumentService.get_tenant_id(id)
|
||||
if not tenant_id:
|
||||
return get_data_error_result(retmsg="Tenant not found!")
|
||||
@ -284,12 +283,13 @@ def change_parser():
|
||||
if doc.parser_id.lower() == req["parser_id"].lower():
|
||||
return get_json_result(data=True)
|
||||
|
||||
e = DocumentService.update_by_id(doc.id, {"parser_id": req["parser_id"], "progress":0, "progress_msg": "", "run": 1})
|
||||
if not e:
|
||||
return get_data_error_result(retmsg="Document not found!")
|
||||
e = DocumentService.increment_chunk_num(doc.id, doc.kb_id, doc.token_num*-1, doc.chunk_num*-1, doc.process_duation*-1)
|
||||
e = DocumentService.update_by_id(doc.id, {"parser_id": req["parser_id"], "progress":0, "progress_msg": ""})
|
||||
if not e:
|
||||
return get_data_error_result(retmsg="Document not found!")
|
||||
if doc.token_num>0:
|
||||
e = DocumentService.increment_chunk_num(doc.id, doc.kb_id, doc.token_num*-1, doc.chunk_num*-1, doc.process_duation*-1)
|
||||
if not e:
|
||||
return get_data_error_result(retmsg="Document not found!")
|
||||
|
||||
return get_json_result(data=True)
|
||||
except Exception as e:
|
||||
|
||||
@ -21,7 +21,7 @@ from api.db.services.user_service import TenantService, UserTenantService
|
||||
from api.utils.api_utils import server_error_response, get_data_error_result, validate_request
|
||||
from api.utils import get_uuid, get_format_time
|
||||
from api.db import StatusEnum, UserTenantRole
|
||||
from api.db.services.kb_service import KnowledgebaseService
|
||||
from api.db.services.knowledgebase_service import KnowledgebaseService
|
||||
from api.db.db_models import Knowledgebase
|
||||
from api.settings import stat_logger, RetCode
|
||||
from api.utils.api_utils import get_json_result
|
||||
|
||||
@ -22,7 +22,7 @@ from api.db.services.user_service import TenantService, UserTenantService
|
||||
from api.utils.api_utils import server_error_response, get_data_error_result, validate_request
|
||||
from api.utils import get_uuid, get_format_time
|
||||
from api.db import StatusEnum, UserTenantRole
|
||||
from api.db.services.kb_service import KnowledgebaseService
|
||||
from api.db.services.knowledgebase_service import KnowledgebaseService
|
||||
from api.db.db_models import Knowledgebase, TenantLLM
|
||||
from api.settings import stat_logger, RetCode
|
||||
from api.utils.api_utils import get_json_result
|
||||
|
||||
Reference in New Issue
Block a user