|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| import base64
|
| import pathlib
|
| import re
|
|
|
| import flask
|
| from elasticsearch_dsl import Q
|
| from flask import request
|
| from flask_login import login_required, current_user
|
| from rag.nlp import search
|
| from rag.utils import ELASTICSEARCH
|
| from api.db.services import duplicate_name
|
| 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, TaskStatus, ParserType
|
| from api.db.services.document_service import DocumentService
|
| from api.settings import RetCode
|
| from api.utils.api_utils import get_json_result
|
| from rag.utils.minio_conn import MINIO
|
| from api.utils.file_utils import filename_type, thumbnail
|
|
|
|
|
| @manager.route('/upload', methods=['POST'])
|
| @login_required
|
| @validate_request("kb_id")
|
| def upload():
|
| kb_id = request.form.get("kb_id")
|
| if not kb_id:
|
| return get_json_result(
|
| data=False, retmsg='Lack of "KB ID"', retcode=RetCode.ARGUMENT_ERROR)
|
| if 'file' not in request.files:
|
| return get_json_result(
|
| data=False, retmsg='No file part!', retcode=RetCode.ARGUMENT_ERROR)
|
| file = request.files['file']
|
| if file.filename == '':
|
| return get_json_result(
|
| data=False, retmsg='No file selected!', retcode=RetCode.ARGUMENT_ERROR)
|
|
|
| try:
|
| e, kb = KnowledgebaseService.get_by_id(kb_id)
|
| if not e:
|
| return get_data_error_result(
|
| retmsg="Can't find this knowledgebase!")
|
|
|
| filename = duplicate_name(
|
| DocumentService.query,
|
| name=file.filename,
|
| kb_id=kb.id)
|
| location = filename
|
| while MINIO.obj_exist(kb_id, location):
|
| location += "_"
|
| blob = request.files['file'].read()
|
| MINIO.put(kb_id, location, blob)
|
| doc = {
|
| "id": get_uuid(),
|
| "kb_id": kb.id,
|
| "parser_id": kb.parser_id,
|
| "parser_config": kb.parser_config,
|
| "created_by": current_user.id,
|
| "type": filename_type(filename),
|
| "name": filename,
|
| "location": location,
|
| "size": len(blob),
|
| "thumbnail": thumbnail(filename, blob)
|
| }
|
| if doc["type"] == FileType.VISUAL:
|
| doc["parser_id"] = ParserType.PICTURE.value
|
| if re.search(r"\.(ppt|pptx|pages)$", filename):
|
| doc["parser_id"] = ParserType.PRESENTATION.value
|
| doc = DocumentService.insert(doc)
|
| return get_json_result(data=doc.to_json())
|
| except Exception as e:
|
| return server_error_response(e)
|
|
|
|
|
| @manager.route('/create', methods=['POST'])
|
| @login_required
|
| @validate_request("name", "kb_id")
|
| def create():
|
| req = request.json
|
| kb_id = req["kb_id"]
|
| if not kb_id:
|
| return get_json_result(
|
| data=False, retmsg='Lack of "KB ID"', retcode=RetCode.ARGUMENT_ERROR)
|
|
|
| try:
|
| e, kb = KnowledgebaseService.get_by_id(kb_id)
|
| if not e:
|
| return get_data_error_result(
|
| retmsg="Can't find this knowledgebase!")
|
|
|
| if DocumentService.query(name=req["name"], kb_id=kb_id):
|
| return get_data_error_result(
|
| retmsg="Duplicated document name in the same knowledgebase.")
|
|
|
| doc = DocumentService.insert({
|
| "id": get_uuid(),
|
| "kb_id": kb.id,
|
| "parser_id": kb.parser_id,
|
| "parser_config": kb.parser_config,
|
| "created_by": current_user.id,
|
| "type": FileType.VIRTUAL,
|
| "name": req["name"],
|
| "location": "",
|
| "size": 0
|
| })
|
| return get_json_result(data=doc.to_json())
|
| except Exception as e:
|
| return server_error_response(e)
|
|
|
|
|
| @manager.route('/list', methods=['GET'])
|
| @login_required
|
| def list():
|
| kb_id = request.args.get("kb_id")
|
| if not kb_id:
|
| return get_json_result(
|
| data=False, retmsg='Lack of "KB ID"', retcode=RetCode.ARGUMENT_ERROR)
|
| keywords = request.args.get("keywords", "")
|
|
|
| page_number = int(request.args.get("page", 1))
|
| items_per_page = int(request.args.get("page_size", 15))
|
| orderby = request.args.get("orderby", "create_time")
|
| desc = request.args.get("desc", True)
|
| try:
|
| docs, tol = DocumentService.get_by_kb_id(
|
| kb_id, page_number, items_per_page, orderby, desc, keywords)
|
| return get_json_result(data={"total":tol, "docs": docs})
|
| except Exception as e:
|
| return server_error_response(e)
|
|
|
|
|
| @manager.route('/change_status', methods=['POST'])
|
| @login_required
|
| @validate_request("doc_id", "status")
|
| def change_status():
|
| req = request.json
|
| if str(req["status"]) not in ["0", "1"]:
|
| get_json_result(
|
| data=False,
|
| retmsg='"Status" must be either 0 or 1!',
|
| retcode=RetCode.ARGUMENT_ERROR)
|
|
|
| try:
|
| e, doc = DocumentService.get_by_id(req["doc_id"])
|
| if not e:
|
| return get_data_error_result(retmsg="Document not found!")
|
| e, kb = KnowledgebaseService.get_by_id(doc.kb_id)
|
| if not e:
|
| return get_data_error_result(
|
| retmsg="Can't find this knowledgebase!")
|
|
|
| if not DocumentService.update_by_id(
|
| req["doc_id"], {"status": str(req["status"])}):
|
| return get_data_error_result(
|
| retmsg="Database error (Document update)!")
|
|
|
| if str(req["status"]) == "0":
|
| ELASTICSEARCH.updateScriptByQuery(Q("term", doc_id=req["doc_id"]),
|
| scripts="ctx._source.available_int=0;",
|
| idxnm=search.index_name(
|
| kb.tenant_id)
|
| )
|
| else:
|
| ELASTICSEARCH.updateScriptByQuery(Q("term", doc_id=req["doc_id"]),
|
| scripts="ctx._source.available_int=1;",
|
| idxnm=search.index_name(
|
| kb.tenant_id)
|
| )
|
| return get_json_result(data=True)
|
| except Exception as e:
|
| return server_error_response(e)
|
|
|
|
|
| @manager.route('/rm', methods=['POST'])
|
| @login_required
|
| @validate_request("doc_id")
|
| def rm():
|
| req = request.json
|
| try:
|
| e, doc = DocumentService.get_by_id(req["doc_id"])
|
| if not e:
|
| return get_data_error_result(retmsg="Document not found!")
|
| tenant_id = DocumentService.get_tenant_id(req["doc_id"])
|
| if not tenant_id:
|
| return get_data_error_result(retmsg="Tenant not found!")
|
| ELASTICSEARCH.deleteByQuery(Q("match", doc_id=doc.id), idxnm=search.index_name(tenant_id))
|
|
|
| DocumentService.increment_chunk_num(doc.id, doc.kb_id, doc.token_num*-1, doc.chunk_num*-1, 0)
|
| if not DocumentService.delete_by_id(req["doc_id"]):
|
| return get_data_error_result(
|
| retmsg="Database error (Document removal)!")
|
|
|
| MINIO.rm(doc.kb_id, doc.location)
|
| return get_json_result(data=True)
|
| except Exception as e:
|
| return server_error_response(e)
|
|
|
|
|
| @manager.route('/run', methods=['POST'])
|
| @login_required
|
| @validate_request("doc_ids", "run")
|
| def run():
|
| req = request.json
|
| try:
|
| for id in req["doc_ids"]:
|
| info = {"run": str(req["run"]), "progress": 0}
|
| if str(req["run"]) == TaskStatus.RUNNING.value:info["progress_msg"] = ""
|
| DocumentService.update_by_id(id, info)
|
| 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!")
|
| ELASTICSEARCH.deleteByQuery(Q("match", doc_id=id), idxnm=search.index_name(tenant_id))
|
|
|
| return get_json_result(data=True)
|
| except Exception as e:
|
| return server_error_response(e)
|
|
|
|
|
| @manager.route('/rename', methods=['POST'])
|
| @login_required
|
| @validate_request("doc_id", "name")
|
| def rename():
|
| req = request.json
|
| try:
|
| e, doc = DocumentService.get_by_id(req["doc_id"])
|
| if not e:
|
| return get_data_error_result(retmsg="Document not found!")
|
| if pathlib.Path(req["name"].lower()).suffix != pathlib.Path(doc.name.lower()).suffix:
|
| return get_json_result(
|
| data=False,
|
| retmsg="The extension of file can't be changed",
|
| retcode=RetCode.ARGUMENT_ERROR)
|
| if DocumentService.query(name=req["name"], kb_id=doc.kb_id):
|
| return get_data_error_result(
|
| retmsg="Duplicated document name in the same knowledgebase.")
|
|
|
| if not DocumentService.update_by_id(
|
| req["doc_id"], {"name": req["name"]}):
|
| return get_data_error_result(
|
| retmsg="Database error (Document rename)!")
|
|
|
| return get_json_result(data=True)
|
| except Exception as e:
|
| return server_error_response(e)
|
|
|
|
|
| @manager.route('/get/<doc_id>', methods=['GET'])
|
| def get(doc_id):
|
| try:
|
| e, doc = DocumentService.get_by_id(doc_id)
|
| if not e:
|
| return get_data_error_result(retmsg="Document not found!")
|
|
|
| response = flask.make_response(MINIO.get(doc.kb_id, doc.location))
|
| ext = re.search(r"\.([^.]+)$", doc.name)
|
| if ext:
|
| response.headers.set('Content-Type', 'application/%s'%ext.group(1))
|
| return response
|
| except Exception as e:
|
| return server_error_response(e)
|
|
|
|
|
| @manager.route('/change_parser', methods=['POST'])
|
| @login_required
|
| @validate_request("doc_id", "parser_id")
|
| def change_parser():
|
| req = request.json
|
| try:
|
| 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.lower() == req["parser_id"].lower():
|
| return get_json_result(data=True)
|
|
|
| if doc.type == FileType.VISUAL or re.search(r"\.(ppt|pptx|pages)$", doc.name):
|
| return get_data_error_result(retmsg="Not supported yet!")
|
|
|
| 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:
|
| return server_error_response(e)
|
|
|
|
|
| @manager.route('/image/<image_id>', methods=['GET'])
|
|
|
| def get_image(image_id):
|
| try:
|
| bkt, nm = image_id.split("-")
|
| response = flask.make_response(MINIO.get(bkt, nm))
|
| response.headers.set('Content-Type', 'image/JPEG')
|
| return response
|
| except Exception as e:
|
| return server_error_response(e)
|
|
|
|
|